The Accessibility of Parkinson's Disease Medication in Kenya: Results of a National Survey

Abstract

Background

There is a dearth of knowledge about the availability and affordability of the different drug treatments for Parkinson's disease (PD) across sub‐Saharan Africa (SSA). We aimed to determine the availability and affordability of drugs for treating PD in Kenya.

Methods

A facility‐based survey was conducted in selected medicine outlets (pharmacies) in what were formerly the headquarter towns of the eight provinces of Kenya. We used the World Health Organization/Health Action International methodology to obtain data for drugs used to treat PD. Unit price for each drug was obtained.

Results

Forty‐eight outlets were visited in total, six in each of Kenya's eight provinces. Levodopa (L‐dopa) was available in only 24 (50.0%) outlets. Only one public pharmacy sold l‐dopa (14 were private and nine were other types of outlet). Ergot‐derived dopamine agonists (DAs) and anticholinergics were available in 37 and 35 outlets, respectively. Monoamine‐oxidase inhibitors, non‐ergot‐derived DAs, and catechol‐O‐methyl transferase inhibitors were available in four, two and zero outlets, respectively. Mean cost of 100 l‐dopa tablets was $48.2, though costs varied widely (range, $28.2–$82.4). Only five outlets considered l‐dopa affordable, all of which sold 100 tablets for less than $31.

Conclusion

There is a lack of availability of PD drugs in Kenya, particularly in public pharmacies, where costs are generally lower. Few pharmacists consider the drugs available to be affordable. If PD is to be effectively managed in Kenya, then strategies are needed to increase the availability and affordability of medication.

Little is known about the epidemiology of Parkinson's disease (PD) in many countries of sub‐Saharan Africa (SSA).1 Data from Tanzania suggest that it is less prevalent than in many high‐income world regions.2, 3, 4 This is likely to be, in part, a result of relatively high mortality rates in those with PD in countries where diagnosis is rare and effective treatment even rarer.5, 6 It is estimated that up to half the people in the world with PD are not diagnosed and therefore not treated.4, 7 Disability levels of people living in the community in SSA are high, with a high burden on carers and other family members.8, 9, 10

Even if people are diagnosed, there is still a major issue about the availability, affordability, and sustainability of drug treatment. The World Health Organization (WHO) model list of essential medicines for PD includes levodopa (L‐dopa) and anticholinergic drugs, but none of the newer drugs, such as non‐ergot dopamine agonists (DAs), catechol‐O‐methyl transferase (COMT) inhibitors, or monoamine oxidase (MAO) inhibitors.11 In SSA, availability of those drugs on the essential medicines list appears to be limited and many drugs may be unaffordable to many people living in SSA countries. The cost of a month's supply of medication may be more than the combined household income. There is also likely to be a significant discrepancy between rural and urban areas, with those in the most rural areas being least likely to have access to drugs. An Ethiopian study reported that over 90% of patients were on l‐dopa treatment, though this is perhaps unsurprising given that the study population were hospital patients.12 This compares to the rural, community‐based, Tanzanian prevalence study where only 15% of patients had ever been treated.4 The aim of this study was to determine the availability, price, and affordability of drugs used to treat PD in Kenya.

Materials and Methods

Ethics

The study was a survey of pharmacists and did not involve patients. The ethical review committee for the University of Nairobi and Kenyatta National Hospital (KNH‐UON ERC) approved the study. The purpose of the study was explained to pharmacists/pharmacist technicians or officers in charge of the medicine outlet at all of the selected facilities before they were invited to participate. Participation was entirely voluntary.

Study Design

This was a survey of six medicine outlets in each of what were formerly the headquarter towns/cities of all eight provinces in Kenya. We used the WHO/Health Action International (WHO/HAI) methodology to obtain data for drugs used to treat PD.13 Unit price for each drug was obtained and affordability was based on the approximate day's wage of the lowest paid government worker, after deduction of basic living expenses, to the unit price of each medicine available.

The monthly salary of the lowest paid government worker was determined based on the national minimum wage in Kenya of 13,674 Kenyan shillings (KSh), with a net pay of 11,222 KSh after tax. Based on statistical models of the cost of living in Kenya (numbeo.com), the percentage of monthly salary spent on basic living expenses is 25.2% on rent, 15.7% on transportation, 33.7% on food, 9.8% on utilities, and 8.6% on sport and leisure activities. This leaves 7% of net salary for other items, such as health care costs. Based on the minimum wage, this 7% equates to 785.54 KSh, or $9.24 US, that could, in theory, be set aside for PD medication per month. Therefore, to determine affordability of PD medicines in Kenya, we compared the approximate day's wage, after all basic expenses were deducted, to the unit price of each medicine available.

Setting and Selection of Pharmacies Surveyed

Pharmacies were divided into three types:

1. Public sector medicine outlets: These were run by the government and generally located in government‐run buildings, such as hospitals, clinics, and health centers.

2. Private sector medicine outlets: These included licensed retail pharmacies and licensed drugs stores. They were operated as private businesses by the owners.

3. Other medicine outlets: These were neither public nor private sector, but run by other organizations, such as charities and nongovernmental organizations (NGOs), health facilities run by religious organizations, private hospitals, and dispensing doctors.

In each town/city, all medicine outlets (pharmacies or dispensaries) within a 3‐hour drive of the main public hospital (e.g., general hospital/regional hospital/tertiary hospital) were identified from national lists. All outlets were then grouped according to whether they were a private, public, or other outlet and randomized. The aim was to visit two pharmacies of each type in each province based on their order on the randomized list. In some cases, none or only one outlet was listed for a particular province and so other types of outlet were visited.

Data Collection

All data were collected by the study research associate (J.M.) from September to November 2014. Data were collected from either the head pharmacist, the owner/manager of the outlet, a pharmacy technician, or a doctor if the outlet was run by a dispensing doctor. Data were collected on a standardized pro forma, which included questions regarding the type, dose, and price of all drugs for PD sold. We also asked whether the interviewee considered the drugs available were affordable to most people in the local area, whether there were any issues with drug supply, whether they were restricted to supplying drugs from the WHO essential medicines list, what were the barriers to people with PD in the town/city receiving appropriate antiparkinsonian medication, and what other factors might limit the availability of treatment for people with PD. The study focused on branded pharmaceutical products whose names were identified on the drug list and lowest priced generically equivalent products.

The research associate assisted respondents in understanding and filling in the questionnaire when needed and checked for completeness, accuracy, and uniformity of the collected data. Data were coded and entered into a spreadsheet for analysis.

Statistical Analysis

Statistical analysis was conducted using IBM SPSS statistics software (version 21; IBM Corp., Armonk, NY). Data were summarized using simple descriptive statistics (range, mean, standard deviation [SD], and frequency, as appropriate).

Results

We collected data from 48 pharmacies. Thirty‐six were contacted in person (75.0%), nine by phone (18.8%), and three by e‐mail (6.3%). In 43 cases, the information was obtained from a pharmacist (89.6%), in three cases from the outlet owner/manager (6.3%), in one case from a pharmacy technician (2.1%), and in one case from a doctor (2.1%). Table 1 summarizes the nature and location of the pharmacies visited. We collected data on whether pharmacies had treated anyone with PD, but did not collect data on exact number of patients seen. Three of fifteen (20.0%) public pharmacies, 2 of 19 (10.5%) private pharmacies, and 1 of 14 “other” pharmacies (7.1%) had not come across anyone with PD. Therefore, overall 42 of 48 (87.5%) pharmacies had seen patients with PD. Figure 1 summarizes responses to a question regarding barriers to accessing antiparkinsonian medication. The single most important barriers are summarized in the figure. Affordability and availability were thought to be the most common issues with regard to people with PD getting appropriate medication (cited by 14.6% and 12.5%, respectively) of those surveyed. When asked about other problems for people with PD (other than the reason already given), a further three respondents (6.3%) cited affordability and two (4.2%) availability. Fourteen respondents (29.2%) cited awareness of PD as a problem and five (10.4%) cited limited health care provision (lack of health care workers, lack of diagnosis or treatment, or lack of specialist health care facilities).

Table 1.

Description of pharmacies surveyed in each of the eight provinces of Kenya

Region	Regional Population	Town	Population of Town	Type of Pharmacies Surveyed
				Public	Private	Othera
Central Province	4,383,743	Nyeri	693,558	2	2	2
Coast Province	3,325,307	Mombasa	939,370	2	4	0
Eastern Province	5,668,123	Machakos	1,098,584	2	2	2
Nairobi Province	3,138,369	Nairobi	3,138,369	2	2	2
North Eastern Province	2,310,757	Garissa	623,060	2	2	2
Nyanza Province	5,442,711	Kisumu	968,909	1	3	2
Rift Valley province	10,006,805	Nakuru	307,990	2	2	2
Western Province	4,334,282	Bungoma	1,375,063	2	2	2

^a“Other” pharmacies were neither public nor private sector, but run by other organizations, such as charities and NGOs, health facilities run by religious organizations, private hospitals, and dispensing doctors.

Figure 1.

Availability and affordability of antiparkinsonian medication in Kenya.

All public pharmacies and almost two‐thirds of private pharmacies had experienced problems in PD medication supply. Interestingly, 35.4% of those surveyed felt there were no barriers to PD patients receiving medication and 37.5% felt there were no limitations to the management of people with PD in Kenya.

Data on the availability and cost per 100 tablets of various classes of antiparkinsonian drugs are described in Table 2. l‐dopa was available in only 24 (50.0%) outlets. Nineteen outlets had only one preparation of l‐dopa available, six outlets had two different preparations available, and one outlet had three preparations available. l‐dopa as cocareldopa 10/100 was available in 20 outlets, cocareldopa 25/100 in three outlets, and cocareldopa 25/250 in nine outlets. No other preparations of l‐dopa (e.g., cobeneldopa) were available. The cost of 100 l‐dopa tablets varied widely depending on the type of outlet and the preparation (see Table 2). Data on the availability and affordability of l‐dopa medication by town/city are shown in Figure 2. Only four outlets (both of those in Eastern, one in Central, and one in Nyanza Province) considered l‐dopa affordable, and all of these sold 100 tablets at less than $31.

Table 2.

Availability, cost, and affordability of medication for PD in Kenya

	Available	Mean Cost per 100 Tablets in U.S. Dollars (Range, standard deviation)	Affordable
l‐dopa (%)
Public (n = 15)	1 (6.7)	25/250 mg: 52.94	0
Private (n = 19)	14 (73.7)	10/100 mg: 38.53 (28.24–38.82, 3.147)	2 (10.5)
		25/100 mg: 47.06 (41.18–52.94, 5.880)
		25/250 mg: 55.68 (51.76–62.35, 5.804)
Other (n = 14)	9 (64.3)	10/100 mg: 50.14 (28.24–38.82, 25.137)	2 (22.2)
		25/250 mg: 68.63 (52.94–82.35, 14.803)
Ergot‐derived DAs (%)
Public (n = 15)	7 (46.7)	1.25 mg: 35.29 (at two outlets)	0
		2.5 mg: 41.18 (at five outlets)
Private (n = 19)	17 (89.5)	0.5 mg: 82.28 (70.36–94.12, 11.880)	3 (17.6)
		1.25 mg: 35.29 (one outlet)
		2.5 mg: 50.19 (29.41–152.94, 30.488)
		5 mg: 55.88 (29.41–82.35, 37.434)
Other (n = 14)	13 (92.9)	0.5 mg: 50.12 (one outlet)	2 (15.4)
		2.5 mg: 43.22 (29.41–89.41, 15.783)
		5 mg: 29.41 (one outlet)
Non‐ergot‐derived DAs (%)
Public (n = 15)	0	–
Private (n = 19)	2 (10.5%)	0.125 mg: 30.59	1 (50.0)
		0.25 mg: 58.82 mg
		1 mg: 123.50 mg
Other (n = 14)	0	–
COMT inhibitors (none available)
MAO inhibitor (type B) (%)
Public (n = 15)	1 (6.7)	5 mg: 35.29	0
Private (n = 19)	1 (5.3)	5 mg: 12.94	0
		10 mg: 47.06
Other (n = 14)	2 (14.3)	5 mg: 16.81 (16.00–17.70, 1.195)	0
Anticholinergics (%)
Public (n = 15)	9 (60.0)	5 mg: 1.91 (1.20–2.40, 0.608)	9 (100)
Private (n = 19)	16 (84.2)	2 mg: 11.76 (one outlet)	16 (100)
		5 mg: 5.65 (1.80–23.50, 6.131)
Other (n = 14)	10 (71.4)	5 mg: 4.89 (2.40–10.10, 2.529)	10 (100)
Atypical antipsychotics (%)
Public (n = 15)	2 (13.3)	50 mg (quetiapine): 23.53	1 (50.0)
		100 mg (quetiapine): 47.07
Private (n = 19)	7 (36.8)	50 mg (quetiapine): 31.37 (29.40–35.30, 3.395)	2 (28.6)
		100 mg (quetiapine): 47.07 (37.80–64.70, 10.409)
		200 mg (quetiapine): 94.12
Other (n = 14)	2 (14.3)	100 mg (quetiapine): 37.84 (at both outlets)	0
Cognitive enhancers (%)
Public (n = 15)	0	–
Private (n = 19)	4 (21.1)	5 mg (donepezil): 44.71 (42.35–47.06, 2.355)	0
		5 mg (Aricept): 174.59
Other (n = 14)	1 (7.1)	5 mg (donepezil): 42.35	0

Figure 2.

Availability and affordability of l‐dopa medication in each town/city.

Ergot‐derived DAs and anticholinergics were available in 37 and 35 outlets, respectively. MAO inhibitors (type B), non‐ergot‐derived DAs, and COMT inhibitors were available in four, two, and zero outlets, respectively.

Discussion

Our study is the first to look at access to antiparkinsonian medications in SSA. l‐dopa preparations were available in 50% of all pharmacies, though only one of these was a public, government‐run, dispensary. Furthermore, only four interviewees considered l‐dopa to be affordable. However, the picture is complicated when one considers the dosages available. Only three pharmacies had l‐dopa available in a 4:1 formulation of l‐dopa to carbidopa, the dopa‐decarboxylase inhibitor, the formulation typically used in high‐income countries. Formulations that contain a lower proportion of dopa‐decarboxylase inhibitor (e.g., 10:1) may result in noncompliance with medication because of side effects. Likewise, without the lower‐dose preparations of l‐dopa/carbidopa being available, initiating therapy, especially in older patients, may be difficult. Ogunniyi reported that the treatment of PD in developing countries commonly starts with anticholinergic medications, owing to their relatively low cost and wide availability.14 Indeed, in our study, anticholinergic drugs were available in 35 of the 48 outlets and were seen as affordable by all interviewees. This opinion is supported by consideration of our calculation of the monthly disposable income ($9.24 US) of a government worker on the minimum wage. Based on this estimate, only anticholinergics would be considered affordable.

Antiparkinsonian medications used routinely in high‐income countries, such as nonergot DAs and MAO inhibitors (type B) were only available in two and four outlets, respectively, and were generally seen as unaffordable. COMT inhibitors were not available in any outlet. Atypical antipsychotics and cognitive enhancers were not readily available outside private pharmacies. Atypical antipsychotics can be used to treat behavioral problems associated with cognitive impairment and confusion in PD, without too much negative impact on motor symptoms.15 Cognitive enhancers can be used to improve memory and reduce hallucinations, which are common problems in late‐stage disease.16 Although some cognitive enhancers are now off‐patent, and so potentially more affordable, they were only available in five outlets.

One major concern about starting drug treatment for patients at a later stage of PD, who have previously been drug naïve, is the possibility of rapidly inducing severe motor fluctuations and dyskinesia. To avoid this, it is generally recommended to start treatment at a low dose and titrate up to the optimal dose slowly.5 Monthly local supervision by a PD nurse specialist during the initiation of drug therapy is recommended and can allow the safe introduction of drug therapy, even in late‐stage disease.5, 17

Although numbers were small because of a general lack of availability of most medications, there did not seem to be any major differences in the cost of drugs between the different pharmacies. Slightly higher costs in private pharmacies were generally the result of stocking of more expensive drugs. Only for anticholinergic medication did the drugs appear to be appreciably less expensive in public pharmacies.

There are few previous reports of the affordability of PD drugs in other low‐ and middle‐income countries, and none from SSA. A 2006 study from China found the average annual direct medical costs of PD to be $519 US, of which $323 US (62.2%) was drug costs.18 The mean overall annual financial cost of PD was around half of the mean annual income. A study in India drew similar conclusions, with people with PD spending between 16% and 42% of the average gross national income on medication.19 In both India and China, as in Kenya, the cost burden for those on low incomes is likely to make the use of regular, optimal PD medication impossible in the absence of government or charitable intervention.

Increasing affordability of medication in many low‐ and middle‐income countries is a complex issue involving numerous stakeholders, such as patients, clinicians, pharmacists, pharmaceutical companies, charities, health care providers, and governments. There needs to be a willingness on the part of all stakeholders to reduce prices. However, training of local health care workers and education of patients to increase accessibility to medication that is currently available and affordable is a realistic aim. The approach proposed by the WHO mental health Gap Action Programme initiative may have some applicability to PD patients in this setting.20

The main limitation of our study is that it is based on a survey of pharmacies conducted within a 3‐hour drive of the provincial capitals and may not reflect the situation in more isolated rural communities. Nevertheless, we have covered all eight provinces of Kenya, and as such our data should be nationally representative of the situation in Kenya.

Conclusions

In SSA, previous data suggest that most patients with PD are undiagnosed.4 The current study points to a lack of available and affordable drug treatment for those who are diagnosed in Kenya. This is likely to be true of most countries in SSA. Raising awareness and enhancing training for health care workers to refer possible cases of PD for diagnosis and treatment will have little impact on their care until antiparkinsonian medication becomes more accessible and affordable to most people.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.

J.M.: 1A, 1B, 1C, 3A, 3B

C.L.D.: 1A, 1B, 3B

W.K.G.: 1C, 2A, 2B, 2C, 3A, 3B

J.H.: 3B

R.W.W.: 1A, 1B, 3B

Disclosures

Funding Sources and Conflicts of Interest: J.M.'s time to carry out data collection was funded by the African Task Force of the International Parkinson and Movement Disorder Society. The authors report no conflicts of interest.

Financial Disclosures for previous 12 months: The authors declare that there are no disclosures to report.