Ageing is an inevitable phenomenon and can be accompanied by declining health with consequent accumulation of chronic illnesses including cardiometabolic diseases, neurocognitive disorders and musculoskeletal conditions. Having multiple chronic conditions can culminate in a high burden of chronic medication use. Adherence to this medication can pose challenges. In South Africa, the number of older persons aged 60 years and above is estimated to be 5.6 million, 9.2% of the entire population.¹ Although 65 years is the definition of geriatric worldwide, in South Africa and the rest of Africa, the World Health Organisation uses the definition of 60 years as an older person.¹

Medication adherence is defined as the extent to which patient behaviours in taking medication and executing dietary and lifestyle changes align with recommendations from healthcare providers.²

Barriers to medication can be divided into categories comprising patient factors, medical factors, healthcare provider factors, healthcare systems factors and socio-economic factors.^3,4,5 Challenges to adherence are a hindrance in the treatment of chronic illnesses in the older population. Non-adherence can be either intentional or non-intentional. Intentional non-adherence refers to the patient taking medication differently from the prescriber’s recommendations. Unintentional non-adherence refers to the patient wanting to take medication as prescribed but being unable to do so because of factors such as poor resources, a lack of understanding, forgetfulness, concerns about effectiveness, side effects and physical barriers such as poor dexterity and swallowing difficulties.^3,6

Various subjective tools using questionnaires and scales are available to assist in the measurement of medication adherence. The Medication Adherence Rating Scale (MARS) is used to assess adherence based on three dimensions: medication adherence behaviour, attitude towards taking medication and attitudes to the negative side effects of prescribed medications.⁷ The higher the score, the better the adherence. The Adherence Barrier Questionnaire (ABQ) is a practicable, reliable and validated instrument for identifying patient-specific barriers to medication-related adherence. It is a 15-item scale comprising the barriers assessing intentional adherence, medication and healthcare system-related adherence and unintentional adherence.⁸

Polypharmacy can either be defined in terms of being prescribed more medication than necessary or in terms of the number of medications taken, usually more than five. These medications include prescribed medication, over-the-counter medication and nutraceuticals.⁹ The prevalence of polypharmacy in a South African hospital in the free state was very high at 84.3%.¹⁰ Apart from the potential of non-adherence, polypharmacy increases the risk of drug interactions and adverse drug reactions.¹¹ Excessive polypharmacy is defined as the use of 10 or more medications, the consequences of which may lead to poor health outcomes.¹² Appropriate pharmacy occurs when multiple medications are used in such a way that each one has a clear indication with minimum risk of harm.¹³ The hallmark of appropriate pharmacy is rational prescribing whereas polypharmacy can be harmful when not properly managed. With polypharmacy, the pill burden is often high and can lead to poor adherence.¹⁴ Inappropriate pharmacy pill burden is minimised by optimising treatment plans.

Multimorbidity refers to the presence of two or more chronic conditions, and together with polypharmacy, is one of the geriatric syndromes.^15,16 Multimorbidity and polypharmacy are closely related as multimorbidity often requires the use of numerous medications. Both polypharmacy and multimorbidity may impact a number of other geriatric syndromes such as frailty.¹⁷

In the paucity of published studies on adherence to medication in geriatric patients at medical outpatient departments (MOPD) in South Africa, we aimed to investigate medication adherence and the barriers to this adherence in geriatric patients at the Helen Joseph Hospital MOPD.

A total of 157 patients were approached for this study and 27 were excluded as they did not provide consent. A total of 130 patients participated in the study.

According to the World Health Organization, adherence rates in developing countries are likely worse than the 50% adherence rates seen in developed countries.¹⁹ This is important because non-communicable diseases contribute a large proportion to the global disease burden.¹⁹ Older people are more likely to have multiple chronic conditions, a higher rate of polypharmacy and a higher prevalence of non-adherence because of the high pill burden, adverse drug reactions and drug–drug interactions. Non-adherence itself is associated with adverse health outcomes such as hospitalisation, emergency department visits and mortality.²⁰

The demographic breakdown of our study showed that the majority of participants were female, which may give some insight into the health-seeking behaviour of patients in our hospital’s drainage area. Just over 40% of participants had only a primary level of education, which may reflect the poorer socio-economic status of our patient population; however, this did not have an impact on adherence in our study. Previous studies showed that marital status influenced medication adherence;²¹ however, this was not the case in our study.

In our study, the self-reported adherence rate was very high at 95%. This contrasts with other studies conducted in Africa. A study conducted in North-West Ethiopia showed that 25.4% of participants of all ages had low adherence. Unlike our study group, this group of patients all had type 2 diabetes. Our results showed that patients living with type 2 diabetes were less compliant than their non-diabetic counterparts.²² In a large Chinese study, the prevalence of non-adherence in geriatric outpatients visiting tertiary centres was as high as 31.8%.²³ In India, the prevalence of poor adherence was only 2%, which is very similar to our study.²³ The differences in the prevalence rates in all of these studies could be because of the variety of different tools used to assess adherence and differences in the population groups and the study settings. Our hospital is a tertiary institution and cannot be directly compared to non-tertiary level hospitals. Almost 95% of our participants acknowledged that the medication helped only if they were taken regularly as recommended. Most of our participants took their medicines regularly at a fixed time.

Factors that influence adherence can be divided into those that are modifiable and those that are non-modifiable. Potentially modifiable barriers include disease-related knowledge, health literacy, social support and polypharmacy.^24,25 These are the barriers that should be given priority as this is where meaningful change can be achieved.

Despite the high rate of adherence, 96% of the participants in our study had at least one barrier to adherence. These barriers included patient factors, medication factors, healthcare provider factors, health systems factors and socio-economic factors. Transport issues and stock availability were some of the issues highlighted in this study. A qualitative study conducted in Iran showed a number of themes relating to poor adherence.²⁶ These included medication errors, forgetfulness, cultural factors, socio-economic factors and fear of complications.²⁶ A scoping review that included barriers to adherence in low- to middle-income countries such as ours showed that lack of knowledge about prescribed medication, negative attitudes and negative beliefs influenced adherence.²⁷ The use of traditional medicines is common in South Africa, and this has been associated with poor adherence in other studies, particularly in people living with diabetes.²² Reasons for this may include patients’ attitudes and the preference to consult with traditional healers but the relationship is not completely understood. This does however highlight the importance of asking our patients whether they use traditional medications, and if so, how they use them.^22,28

Medication-specific barriers are a common theme among studies looking at non-adherence. In our study, 39% of participants were worried about the side effects of their medication and 49.23% had stopped at least one medication because of patients experiencing side effects. Worryingly, 22% of participants believe medications are poisonous and should be avoided if possible. We extrapolate that this may be because of patients experiencing side effects or may be because of cultural factors. Attitudes to different medications may be influenced by socio-demographic and cultural factors.^29,30 The use of traditional medicine was shown to be independently associated with poor adherence in a study focussing specifically on sub-Saharan countries.³¹ Hypertensive patients in America reported medication side effects, pill size and difficulty swallowing pills as some of their barriers to adherence.³² Some studies have shown that complicated regimes are a barrier to adherence³³ and this is of particular importance in the geriatric population where forgetfulness is a major contributor to non-adherence. In fact, our study showed that 59% of patients had forgotten to take their prescribed medication at least once in the preceding month.

Certain healthcare-related factors such as the availability and accessibility of medications were not statistically significant in this study, but they still deserve a mention because stock shortages are relatively common in our setting and the financial burden of transportation impacts adherence.

Our study showed that having a good relationship with the healthcare provider meant adherence was more likely. Ninety-eight per cent of our cohort fully trusted their health care practitioner and understood what had been explained to them. In our study, most patients knew what their medication was for and trusted their healthcare provider. Other studies corroborate this finding – in a primary health care setting, adherence was improved by good communication with the treating doctor, social support and knowledge about the disease and treatment thereof.³²

One of the most common geriatric syndromes is multimorbidity or the presence of two or more chronic health conditions. In our study, the most common diseases seen included type 2 diabetes mellitus, hypertension, dyslipidaemia and congestive cardiac failure. These conditions seem to be the most common conditions seen in other studies exploring adherence in older patients.³⁴ In our hospital setting, organ-specific diseases are often seen at dedicated sub-specialist clinics and not at MOPD. Certain comorbidities may be underrepresented in our study.

Our study showed a 53% prevalence of polypharmacy, a lot higher than the prevalence in a study conducted in a dedicated geriatric centre in Nigeria.³⁵ In that study, polypharmacy was associated with factors leading to non-adherence such as visiting many different clinics and intentionally missing medication doses because of high pill burden. In Durban, a study conducted at a regional hospital revealed that the average number of prescribed medicines was 12.¹¹ In our study, only 2% of participants were prescribed 10 medications or more. Moreover, in the Durban study, the highest prevalence of polypharmacy was found in the 60–65-year age group. In an American study, the average number of medications was 5.7 prescription drugs.³³ A recent systematic review published in the Lancet showed the prevalence of polypharmacy ranged from 2.6% to 86.6%. Although our study was not primarily focussed on polypharmacy, the high rate of polypharmacy may increase adverse drug reactions and drug–drug interactions which in turn lead to further polypharmacy and the creation of prescription cascades.^36,37 In older adults with multimorbidity and polypharmacy, medication side effects can be mistaken for new conditions. This leads to further prescribing which can cause poor outcomes and further drug-drug interactions.³⁶

Medication adherence is of paramount importance in the management of chronic conditions. Multimorbidity and polypharmacy are common challenges, and as such, adherence in this population can be problematic. Our study showed a high rate of adherence, but despite this, the majority of patients had at least one barrier to medication adherence. The main factor that improved adherence was a good patient–doctor relationship. Perceptions of medications being poisonous need to be addressed in our patient population.