Most populations now have a longer life expectancy than before, and so the prevalence of chronic diseases is increasing [1]. Because chronic diseases are incurable, with some diseases requiring complex regimen or incurring high medical expenses, some patients use herbal medicines [2], which are defined as “herbs, herbal materials, herbal preparations and finished herbal products that contain parts of plants as active ingredients, or other plant materials, or combinations” [3]. Previous studies have shown various prevalence of herbal medicine usage in the elderly, ranging 12.0%–97.4% [4,5,6,7,8,9,10] and depending on the study population and setting. The usage of herbal medicines is common within the Thai population, due to a positive attitude and knowledge about herbal medicines relayed from generation to generation. Thai traditional herbal medicines are legal and accepted, with some herbal medicines being included in the Thai national drug lists. Moreover, these herbal medicines are easy to access, via multiple sources, without a doctor's prescription, such as through a pharmacy or websites. A study indicated that 97.4% of elderly Thai people reported an experience of herbal medicine usage [10]. Being an elderly woman, having higher education, having a good income, and an increasing number of underlying diseases is associated with higher prevalence of herbal medicine usage. By contrast, very advanced age was associated with a lower prevalence of herbal medicine usage [4,5,6, 8].
Common conditions associated with herbal medicine usage are stroke, cancer, and arthritis [4]. The perceptions of herbal medicine use are reduced medical expenses, that they cure diseases, relieve symptoms, and provide good health [10]. However, herbal medicines can be sold and marketed without safety and efficacy profiles, which are required for pharmaceutical drugs. Although some herbal medicines have scientific evidence for their efficacy, such as gingko, which is possibly effective for the treatment of dementia, other herbal medicines are associated with adverse events. Adverse drug events, such as nephrotoxicity, hepatotoxicity, or carcinogenicity, can be caused by the active ingredients of the constituent herbs [2, 7, 11] and contaminants in herbal medicines, such as heavy metals [12], or by interactions between herbal and conventional medicines [5,6,7, 13, 14]. These adverse events may have a higher chance of occurrence in the elderly because of their lower body capacity, high levels of comorbidities, and the number of medications taken. Most users are not aware of the problem of drug interactions [15].
For use of safe herbal medicine, elderly people should consider several factors before deciding to use a herbal medicine, such as, but not limited to, the 4 principles to consider before using complementary and alternative medicine, indicated by the Faculty of Medicine Ramathibodi Hospital, Mahidol University, namely, product reliability, safety, efficacy, and cost–benefit effectiveness [16]. Less than half of elderly users of herbal medicine consulted health care providers regarding herbal medicine use [5, 8, 13], and only 16.6% of users were asked about their herbal medicine usage by their physicians [9].
Although many studies have reported the prevalence and factors related to herbal medicine usage, only a few have been conducted in Thailand and even those in limited settings. These studies have evaluated patient concerns and physician's attention to herbal medicine usage in the elderly. Therefore, the present study had the primary objective of determining the prevalence of herbal medicine usage in the elderly attending a primary care unit (PCU) of a tertiary teaching hospital in Hat Yai, Songkhla province, Thailand; with secondary objectives being to determine factors associated with use, reasons for use, principles considered before use, perceived effects, and history of consultation with medical professionals concerning herbal medicine use.
The present study was approved by the Ethics Committee of the Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla province, Thailand (REC. No. 62-222-9-1). The present study was conducted in compliance with the contemporary revision of the Declaration of Helsinki and the International Conference on Harmonization in Good Clinical Practice. All participants signed informed consent forms after reading the participant information sheet and having it explained to them. This cross-sectional study was conducted in the PCU of Songklanagarind Hospital, from September 2 to 20, 2019. The PCU provides primary health care services for adult and elderly patients within Songklanagarind Hospital, a tertiary care, university teaching hospital of Prince of Songkla University in southern Thailand.
We included elderly patients (aged ≥60 years), who attended the PCU of Songklanagarind Hospital during the study period, with any complaint, who were able to communicate in Thai, and who consented to participate in our research. We excluded patients who required emergency treatment such as a hypertensive crisis. Sample size was calculated based on the primary objective of this study, by using a population proportion formula, n = Z21−α/2 p(1−p)/d2; we used the prevalence of current usage of herbal medicines in the elderly who visited the geriatric clinic, at a tertiary care hospital in Bangkok, Thailand, for the power calculation, being 15.0% [13], 95% confidence interval (CI), a precision error of 5%; 196 participants were required and we added a further 10% to account for potentially missing data. We enrolled patients who were compatible with the eligibility criteria, by convenience sampling method; then we excluded questionnaires with incomplete information, providing data from a remaining total of 204 patients for analysis.
We used questionnaires to assess the primary outcome: “herbal medicine usage,” in which the answers were divided into 3 groups (never used, history of usage, but stopped using for more than 1 year, and current usage, which means a history of herbal medicine use within 1 year) as well as related factors, which were derived from a review of the literature. In addition, we added information about details of herbal medicine usage such as product name, reason and person who recommended their use, health results after using, and experience of physician interviews. The data collection form was sent to 3 expert investigators to determine an Item Objective Congruence (IOC) index score for each question. The results showed an IOC < 0.5 for 2 questions, which were corrected as recommended until IOC ≥ 0.5 for all questions was reached before use. All researchers were trained for standardization and we conducted a pilot study in elderly patients at another PCU, which was close to our study setting.
Four researchers searched for patients who fitted the eligibility criteria of our study; then they performed a consent process, by explaining the details of the research to the participants, and obtained written informed consent. Another 4 researchers, who were trained in questionnaire use, directly asked participants the questions, as we had considered the vision problems of some elderly patients. We then confirmed some of the data with the hospital information system.
Data were entered in EpiData (version 3.1, Denmark), with a double-entry basis, and were analyzed using R software (R Core Team 2017). Descriptive statistical analysis was used to report the baseline characteristics of the patients, prevalence, and detail of herbal medicine use. We presented categorical data in terms of percentage, while continuous data were checked for normal distribution, and median with interquartile range (IQR) was used when normal distribution assumption was not met. We used multivariate logistic regression to test the relations between the associated factors and herbal medicine usage. Data are presented as odds ratio (OR), with 95% CIs.
The baseline characteristics of 204 elderly patients are shown in
Patient characteristics (n = 204)
Sex | |
Male | 62 (30.4) |
Female | 142 (69.6) |
Age (years) (median [IQR]) | 69.0 (9.0) |
Education | |
Uneducated | 13 (6.4) |
Educated‡ | 191 (93.6) |
Marital status | |
Couple | 138 (67.6) |
Single | 13 (6.4) |
Widowed/divorced | 53 (26.0) |
Career | |
No | 142 (69.6) |
Yes | 62 (30.4) |
Salary§ | |
No | 52 (25.5) |
Yes | 152 (74.5) |
Underlying diseases | |
No | 9 (4.4) |
Yes | 195 (95.6) |
Type of underlying diseases | |
Dyslipidemia | 145 (71.1) |
Hypertension | 144 (70.6) |
Diabetes mellitus | 64 (31.4) |
Osteoarthritis | 17 (8.3) |
Cardiovascular diseases | 14 (6.9) |
Gout | 10 (4.9) |
Dyspepsia | 9 (4.4) |
Asthma/chronic obstructive pulmonary disease | 9 (4.4) |
Cancer | 7 (3.4) |
Allergic rhinitis | 7 (3.4) |
Number of underlying diseases (median [IQR]) | 2.0 (1.0) |
Number of current medications || per day (median [IQR]) | 3.0 (1.0) |
IQR, interquartile range.
Data are presented as n (%) unless indicated otherwise. IQR, interquar-tile range.
Educated group means patient who had some degree of education.
Salary means fixed regular income from any sources (accept elderly living allowance, a right of all elderly Thai), such as salary from work or monthly money from relatives.
Current medications refers to medications that patients use regularly for treatment of their underlying diseases (not including vitamin supplements or medication to treat acute illness).
Prevalence of herbal medicine usage is shown in
Prevalence of herbal medicine usage (n = 204)
No | 80 (39.2) |
History of use, but stopped after more than 1 year | 23 (11.3) |
Current usage† | 101 (49.5) |
Current usage means having a history of herbal medicine use within the past 1 year.
Reasons for usage, source of influence to initiate use, disclosure to health care providers, and perceived effects after use in patients with a history of herbal medicine usage (n = 124)
Reasons for use† | |
Thought that herbal medicines are interesting | 81 (65.3) |
Herbal medicines may add benefits to conventional medicine | 60 (48.4) |
Health care provider suggested use | 7 (5.6) |
Conventional medicine was ineffective | 6 (4.8) |
Other | 8 (6.5) |
Expected benefits† | |
Cardiovascular system | 80 (64.5) |
Gastrointestinal system | 36 (29.0) |
Respiratory system | 33 (26.6) |
Endocrine system | 28 (22.6) |
Musculoskeletal system | 27 (21.8) |
Other | 60 (39.0) |
Consideration before herbal medicine usage | |
Did not consider any topic before consumption | 98 (79) |
Considered at least 1 of the following topics before consumption† | 26 (21) |
Product reliability | 20 (16) |
Safety | 8 (7) |
Efficacy | 2 (2) |
Cost–benefit effectiveness | 2 (2) |
Source of influence initial usage† | |
Oneself because of their tradition or culture | 25 (20) |
Suggestion by neighbor | 25 (20) |
Suggestion by relatives | 24 (19) |
Advertising media | 20 (16) |
Suggestion by friends | 18 (15) |
Suggestion by health care providers | 12 (10) |
Suggestion by other patients | 9 (7) |
Other | 13 (11) |
Perceived effects after use | |
Beneficial | 95 (77) |
No difference | 23 (19) |
Experienced side effects‡ | 6 (5) |
Drug interaction awareness | |
No | 78 (63) |
Yes | 46 (37) |
Preferences of disclosure to health care providers | |
Would only report if medical professionals asked | 80 (68) |
Would always ask before use | 16 (13) |
Would always report after use | 12 (10) |
Would disclose only if complications occurred | 8 (7) |
Would not disclose about use | 4 (3) |
History of being asked about herbal medicine use by health care providers | |
No | 91 (73) |
Yes | 33 (27) |
Participants can choose more than 1 answer.
Side effects included diarrhea in 3 patients, abdominal discomfort in 2 patients, and dizziness in 1 patient.
About half of the elderly patients had used a herbal medicines within the past 1 year: the name and number of herbal medicine used are shown in
Number and name of herbal medicine usage in current user (n = 101)
Number of herbal medicines used together | |
1 | 63 |
2 | 16 |
3 | 15 |
4 | 4 |
5 | 1 |
7 | 2 |
Name of herbal medicine used† | |
Kariyat | 37 |
Turmeric | 19 |
Bai yanang | 17 |
Laurel | 14 |
Gotu kola | 10 |
Ginger | 10 |
Moringa | 9 |
Garlic | 7 |
Cordyceps | 7 |
Collagen | 7 |
Herbal medicine (name unknown) | 17 |
Other | 80 |
Participants can choose more than 1 answer.
Multivariate analysis for factors associated with herbal medicine usage in elderly patients (n = 204)
Sex | 0.28 | ||||
Male | 21 (34%) | 41 (66%) | 1 | ||
Female | 59 (42%) | 83 (59%) | 0.68 (0.34 to 1.37) | 0.28 | |
Age (median [IQR]) | 68.8 (8.5) | 69.0 (9) | 0.99 (0.95 to 1.04) | 0.90 | 0.90 |
Education | 0.03* | ||||
Uneducated | 9 (69%) | 4 (31%) | 1 | ||
Educated | 71 (37%) | 120 (63%) | 4.12 (1.12 to 15.2) | 0.03* | |
Marital status | 0.36 | ||||
Couple | 51 (37%) | 87 (63%) | 1 | ||
Single | 7 (54%) | 6 (46%) | 0.43 (0.13 to 0.39) | 0.16 | |
Widowed/divorced | 22 (42%) | 31 (59%) | 1.04 (0.51 to 2.14) | 0.91 | |
Career | 0.28 | ||||
No | 58 (41%) | 84 (59%) | 1 | ||
Yes | 22 (36%) | 40 (65%) | 1.46 (0.74 to 2.88) | 0.28 | |
Salary | 0.85 | ||||
No | 21 (40%) | 31 (60%) | 1 | ||
Yes | 59 (39%) | 93 (61%) | 0.94 (0.46 to 1.89) | 0.85 | |
Number of underlying diseases (median [IQR]) | 2.0 (1.0) | 2.0 (1.0) | 0.74 (0.52 to 1.06) | 0.10 | 0.10 |
Number of current medications per day (median [IQR]) | 3.0 (1.25) | 3.0 (1.0) | 1.06 (0.84 to 1.34) | 0.63 | 0.63 |
CI, confidence interval; IQR, interquartile range; OR, odds ratio; LR, likelihood ratio.
Data are presented as n (%) unless indicated otherwise.
More than half of the elderly patients who attended the PCU had been using herbal medicines, especially educated patients. The majority did not consider any principles for safe use of herbal medicine before consumption; more than half did not know that herbal medicines could interact with conventional medicine and would only disclose a history of usage if they were asked by their health care providers.
The participants of this study were mainly women, by a factor of 2, with a median age of 69.0 years, which was comparable with previous studies [10, 17]. About 7% of participants were not educated, although this corresponds to the results of a national survey in Thailand in 2014, which found that almost 10% of elderly Thai had never been formally educated [18]. Almost all patients had underlying diseases, especially noncommunicable diseases that are often currently incurable, resulting in some trying to find a hopeful treatment to cure their disease. In addition, this group had current, conventional medicines to control their underlying diseases. When using herbal medicine, one should be aware of adverse events from herb–drug interactions [5,6,7, 13, 14].
This study shows the prevalence of herbal medicine usage is being close to that by the elderly in Singapore [17], but is higher than that found by studies conducted in Western countries [4,5,6, 9]. This may be because Asian people generally use herbal medicines more commonly than people of other ethnicities [8]. Especially within the Thai population, people have positive attitudes and easy access to herbal medicines. The common reasons for initiating herbal medicine usage are the same as those found by previous studies [5, 8, 9]. We found that herbal medicine usage is significantly higher among educated people than it was in those without formal education, with an OR of 4.12, and this result is in accordance with previous studies [4, 15]. This may be due to educated patients being more concerned about their state of disease, having better health literacy, ease of access to information about complementary and alternative medicine, and higher perception of benefits from herbal medicine [10]. Therefore, they may be more likely to explore a larger range of therapies for the treatment of their symptoms. The 3 most common herbal medicines used by elderly Thai patients were kariyat, turmeric, and bai yanang. Kariyat (
More than half of the participants with a history of herbal medicine usage did not consider any principles of safe use of herbal medicine before consumption. This finding is in accordance with those of a previous study, which recommended that the knowledge of Thai people in using alternative medicine should be improved [20]. The finding that almost 3 quarters of the health care providers did not place any consideration of herbal medicine usage is unexpected for a university hospital, which provides training for medical students and residents. This may be the result of treatment of elderly patients having multiple issues of concern, such as multiple underlying diseases, drug–drug interactions, drug–disease interaction, pharmacokinetics and pharmacodynamic changes, along with drug side effects, so as the physician may not have had sufficient time to explore complementary and alternative medicine usage in the elderly.
The strengths of the present study are that there are only a few studies on this topic that have been conducted in Southern Thailand. Moreover, this study explores the awareness of principles of safe use of herbal medicine before consumption, including discussion with health care providers and physician's attention to herbal medicine usage, because few studies in Thailand have made mention of this point. There are some limitations to this study. First, the question about herbal medicine usage within 1 year might cause recall bias, although an effort was made to allow enough time for participants to recall their history of usage. Additionally, deep details of usage, such as dosage, were not included. Second, due to awareness of vision problems in elderly patients, this study used verbal questionnaires, which may have caused some participants to feel distressed when answering, thus, possibly causing an underestimation of prevalence of usage. To counteract this limitation, we explained to participants that their answers would not be reported individually, and it would in no way affect care received from their health care providers. Third, we did not record herbal and conventional medicine interactions. This is an important point to explain to patients when advising them to start, or stop, using herbal medicine. Finally, this study was conducted in the PCU of a university hospital, wherein the patients included had a higher amount of underlying diseases when compared with participants in a community setting [5]. A greater number of chronic diseases is recognized as associated with the greater use of herbal medicines [4]. Therefore, health care providers should use caution when applying our results to other settings.
The finding of a high prevalence of herbal medicine usage by elderly patients coupled with low consultation rates suggests that health care providers should be more aware of herbal medicine usage and should increase their role of initiating the discussion about herbal medicine usage with elderly patients, so as to avoid problems arising from the use of herbal medicine, such as herb–drug interactions or adverse effects. Extending the study setting to other settings, such as the general community, increasing the sample size to increase the statistical power of the study, and evaluating other topics regarding herbal medicine usage such as frequency and dosage and herb–drug interactions are warranted.