Globally, stroke remains the second leading cause of death and the third leading cause of disability in adults (Zhang et al., 2021; Feigin et al., 2022). In 2019, stroke was responsible for 143 million disability-adjusted life-years and 6·55 millions of deaths (Feigin et al., 2022). In this context, many published studies have concluded that multiple modifiable or non-modifiable factors may increase the risk of stroke occurrences. Here, it is worth mentioning that approximately three-quarters of all strokes occur in persons older than 65 years (Yousufuddin and Young, 2019). Additionally, Gorelick (2019) suggested that to relieve the future global burden of strokes, prevention among older adults would be an important objective.

Stroke survivors are at an elevated risk of having recurrences (Del et al., 2019), which make up 25–30% of all strokes (Hankey, 2014). As recurrent strokes are associated with high mortality and more disabling (Wu et al., 2019), prescribed medications are recommended for secondary prevention of stroke by guidelines (Coutts et al., 2015; Ahmed et al., 2019). Moreover, several sources of evidence indicate that secondary prevention medications tend to reduce the risk of stroke recurrence (Bushnell et al., 2014; Flach et al., 2020; Yeo et al., 2020). For example, it has been conclusively shown that higher adherence to antithrombotic or statin treatments is associated with a decreased risk of stroke recurrence and mortality (Yeo et al., 2020; Rodriguez-Bernal et al., 2021). In the same vein, Toyoda et al. (2019) reported that the combination of cilostazol with aspirin or clopidogrel resulted in a reduced recurrence of ischemic strokes. In addition, a review conducted by Katsanos and Hart (2020) indicated that an important protection against recurrent strokes is provided by the current triad of pharmacologic mainstays for secondary stroke prevention: blood-pressure lowering drugs, statin drugs, and antiplatelet agents. Meanwhile, there is consensus among scientists that non-medication adherence is a potential modifiable risk factor for poor BP control (Boima et al., 2015; Burnier and Egan, 2019), which is itself a well-established and modifiable risk factor for strokes.

However, patient adherence to medication is largely accounted for the effect of medical management to prevent recurrent strokes and other adverse outcomes (Zhang et al., 2021). Medication adherence is generally defined as the extent to which patients take their medication in line with the recommendations of their healthcare provider (Sabaté, 2003). A large and growing body of literature has reported that post-stroke medication adherence and persistence rates are low in stroke survivors (Kim et al., 2020; Ruksakulpiwat et al., 2020; Zhang et al., 2021). These results, therefore, suggest that improving medication adherence among survivors of stroke should be a growing concern to clinicians, healthcare systems, and other stakeholders (e.g., payers).

There is no doubt that a key step in the creation of an appropriate strategy to improve medication adherence is first understanding stroke survivors’ non-medication adherence and its associated factors. Although several studies have shown the prevalence and factors associated with medication adherence among stroke survivors (Pan et al., 2017; Wei et al., 2017; Ruksakulpiwat et al., 2020), they are not specific to older adults. Furthermore, it is important to note that medication adherence is of a particular concern in older persons. A recent systematic review indicated factors negatively associated with adherence in this population: complex regimens with multiple prescribing physicians; problems with medication storage and formulation; and multimorbidity and cognitive impairment (Smaje et al., 2018). On the other hand, non-medication adherence rates among older adults in other chronic diseases (e.g., diabetes mellitus and hypertension patients) have been found to be alarmingly high (Saqlain et al., 2019; Xu et al., 2020). Meanwhile, older adult stroke survivors are likelier to have other chronic medical conditions (e.g., hypertension, history of cardiac-related comorbidities, diabetes, etc.) (Gruneir et al., 2016; Maresova et al., 2019), which means they have to take more medications to meet their broader health needs, thus further challenging their adherence to medications and an increase in the possibilities of worse health-related risks when non-adherence occurs. However, little is known about the prevalence of non-medication adherence and the factors associated with it among older adult stroke survivors, which this study aims to examine among older adult stroke survivors in Chenzou, Hunan, China.

To the best of our knowledge, this cross-sectional study is the first to evaluate medication non-adherence among older adult stroke survivors in China. Our findings showed that medication non-adherence was observed in 61.4% of the 402 sampled patients. Specifically, partial adherence, low adherence, and poor adherence were found to be 37.1%, 13.7%, and 10.7%, respectively. Moreover, the findings in terms of the non-adherence rates in the current study were much higher than those previously published in a meta-analysis of observational studies regarding post-stroke patients. The latter indicated that the overall non-adherence rate to secondary preventative medication among stroke survivors was 30.9%–35.9% (Al et al., 2016; Zhang et al., 2021). A possible explanation for this is that the aforementioned studies included all stroke patients aged over 18 years, while the present study focused on patients over 60 years of age. In this context, Yuvaraj et al. (2019) found that being part of the elderly age group is a determinant of non-adherence to medications after adjusting for possible confounding variables. Older adults are prone to multiple comorbidities and use more medications than their younger counterparts and may therefore present with a higher risk of medication non-adherence (Chiang-Hanisko et al., 2014). Taken together, our findings present concerns about the alarming rate of non-medication adherence among older adult stroke survivors in Chenzhou, Hunan Province, China, which requires further attention. This finding, while preliminary, suggests that strategies to promote medication adherence among this group are urgently needed.

Meanwhile, the educational level was found to be a predictor of medication non-adherence among older adult stroke survivors. According to our findings, patients who had higher educational levels were less likely to be non-adherent as compared to those who had attended only primary school or less. These results are consistent with those of other studies and suggest that higher education levels are associated with adherence (Kirkman et al., 2015; Jin et al., 2016; Bandi et al., 2017). This could be explained by the fact that those with higher educational levels may be less likely to have negative beliefs about medications (Lemay et al., 2018), which would promote medication adherence. Therefore, when seeking strategies to improve medication adherence among stroke survivors, educational-level factors must be considered. Meanwhile, it is worth noting that the average duration of education for stroke patients was significantly lower in patients aged ≥65 years than in patients aged <65 years (Lu et al., 2018). Such findings help shed light on the educational level as an important challenge for medication adherence interventions among older adult stroke survivors.

Among the common concerns, we found that the medication non-adherence of stroke survivors increased with that of total prescription medications per day. These results are congruent with the findings of earlier studies that indicate older adults are more adherent to a simplified medication regime (Jin et al., 2016). A possible explanation of this result is that patients may be less likely to forget to take medicine if the number of pills is low (Napolitano et al., 2016). However, Kim et al. (2020) reported that optimal medication adherence was associated with more prescribed medicines. This rather contradictory result could be attributed to the study including chronic diseases, while our study only considered elderly stroke patients.

Another important finding was that low needs or high concern regarding medication were associated with medication non-adherence. This finding is in agreement with previous studies on medication adherence among elderly people with chronic diseases (Lemay et al., 2018; Park et al., 2018), which revealed that as compared to patients with high needs and low concerns about medication, those with low needs and high concerns had significantly high medication non-adherence. Our results corroborated that of Horne and Weinman (1999b) who suggested that beliefs in medication may offer accurate predictability about adherence. Additionally, concerns about prescribed medications and unawareness of the rationale for treatments were expressed as primary reasons for non-adherence by stroke survivors (Bauler et al., 2014). Hence, for medical professionals, it is of great importance to outline and educate patients on the necessity of treatments rather than simply providing information about medication, while also managing patients’ concerns by sharing the known side effects of prescribed drugs and helping them recognize and cope with side effects to improve their confidence.

Overall, the most important clinically relevant finding was that health literacy is significantly associated with medication non-adherence. Here, health literacy refers to the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to understand and use information to promote, maintain, or improve their health (Ratzan, 2001). According to this finding, lower health literacy has a greater likelihood of being associated with non-medication adherence. This result matches those observed in earlier studies (Lee et al., 2017; Mayo-Gamble and Mouton, 2018). This may be attributable to the fact that low health literacy is associated with less understanding of prescribed medication instructions (Mayo-Gamble and Mouton, 2018) and problems in using preventive services (Stormacq et al., 2019). As health literacy is considered the most important modifiable risk factor of socioeconomic differences in health (šedová et al., 2021), future work should investigate tailored interventions considering health literacy among older adult stroke survivors.

The most important strength of this study is that to the best of our knowledge, it provides invaluable information on the prevalence and factors associated with medication non-adherence by using standardized rating scales among older adult stroke survivors in Chenzhou, Hunan Province, China. Furthermore, our findings may provide a comprehensive picture of medication non-adherence among stroke survivors, which may lay the groundwork for interventions aimed at increasing adherence among this population. Here, it should be noted that our participant population was older adult stroke survivors, who have been frequently left out of studies. Hence, our study addresses an understudied group.

However, this study has limitations that are worth considering. First, information on medication adherence was collected for the previous months prior to the survey, so some degree of recall bias cannot be ruled out. This could lead to inaccurate estimations of the prevalence of medication non-adherence. Additionally, the data relied on self-reported practices of medication adherence, which might have been over- or under-reported by participants. Second, it was limited in scope. Participants were from three tertiary hospitals in Chenzhou, Hunan Province, China, which limits its generalizability to the broader regions in China. Third, this was a cross-sectional study. Therefore, associations between medication non-adherence and risk factors cannot necessarily be considered causal relationships.

Our findings emphasize that medication adherence among older adult stroke survivors is problematic in this sample of study participants from Chenzhou, Hunan Province, China. Specifically, we found that medication non-adherence was significantly associated with educational levels, number of prescription medications to be taken per day, low needs or high concerns for medication, and health literacy. Building on this, we have offered specific recommendations for proposing population-specific medication adherence interventions among older adult stroke survivors. Our findings also suggest that efforts are needed to explore the prevalence and associated factors of medication non-adherence among older adult stroke survivors in other countries.