This is the first study of its kind in the Caribbean to assess the use of medicinal herbs by asthmatic patients attending a specialty care clinic. The findings of this study are instructive as the use of medicinal herbs for self-medication in disease management has far reaching implications on the quality of healthcare delivery [14]. We report a prevalence of 30.4% in our patient sample, which is significantly higher than that in the UK, Denmark, Singapore and in the US [15-18].

Most patients using medicinal herbs relied on the advice of relatives and friends as their sole source of information, as were caregivers of children in a US study [19]. We suggest that this information on the use of medicinal plants could have come from traditional/cultural knowledge, anecdotal evidence or from the greater public awareness through information networks such as the internet on the potential medicinal benefits of herbs. Asthma is an emerging chronic disease in the Caribbean and we suggest that the traditional knowledge in this area may be relatively 'new' and exist in relation to other diseases affecting the respiratory tract, such as cough, the common cold and the flu. This may be one of the reasons for the low prevalence of use of herbs in elderly asthmatic patients, as a strong traditional knowledge may not have existed.

We expected a higher prevalence of herbal use in individuals living in rural areas as these districts are depots for traditional knowledge as was reported in Jamaica where rural respondents used a larger variety of herbs than those living in urban areas [10]. As suggested earlier, we suspect that due to the recent emergence of asthma as a chronic disease in the Caribbean it is reasonable to expect that traditional knowledge in the management of this disease is not strong and our results are indicative of this.

We suspected that employment status could have predicted the use of herbs, however, this was not the case in our study sample. Unemployed patients did not improvise more in their use of herbal remedies than those in other income groups, even though most of the herbs used were relatively common, readily available and cheap. The low socioeconomic status of the majority of the sample may have prohibited both consultation with qualified herbalists and the purchase of imported, processed herbs that would have incurred additional out-of-pocket expense to the patient. What we noted was that there was no difference in the use of herbs across the income ranges and that in fact, patients earning relatively modest annual incomes between $US12,000 and $US19,999 were most likely to use herbs, although this did not reach statistical significance.

Attaining a higher education positively influence the decision to use herbs. We suggest that in the absence of traditional knowledge regarding the medicinal use of herbs for asthma, a higher educational level may predispose an individual to greater access to general knowledge, especially with greater exposure to the internet and other sources of information, and this could be a factor in positively influencing the individual's decision to use medicinal herbs. The availability of scientific evidence-based information on the efficacy of herbs for diverse healthcare problems may be particularly significant in patients with the resources to avail themselves to such information, particularly those with higher educational and income levels. This is particularly true for garlic and Echinacea, which have been extensively researched and furthermore patients with higher educational and income levels would be more likely be at an advantage to access information via literature or on the world wide web regarding the use of these medicinal plants.

Patients using imported, processed, and obviously more expensive herbal medications were on the higher end of the socioeconomic scale and were more likely to afford these medications. It was also observed that garlic and Echinacea were the herbs of choice in patients with higher educational levels. These herbs have a long tradition of use and are widely researched in Europe and North America. The traditional use and strong scientific evidence to support their therapeutic efficacy could be important factors influencing the patient's decision. It has been suggested elsewhere that patients with higher educational levels also tend be more involved in the management of their health; they tend to self-medicate or even suggest to their physicians the course of therapy.

Although one in five patients using medicinal herbs stated that "conventional medicines were not working" as the reason for using this alternative healthcare modality, we noted that asthma severity does not affect the decision to use herbs. In previous studies, poor management was associated with non-compliance with prescribed pharmacotherapy and poor inhaler technique [5,6].

The backyard and home garden were major sources of readily available herbs such as aloes, shadon beni and lemongrass. Wild growing 'weeds' such as shandileer, tulsi, cocoa onion and black sage were also identified. The supermarket was a major source of inexpensive common medicinal herbs such as garlic, ginger and nutmeg. The identification of these medicinal herbs provides an opportunity to investigate West Indian plants used to treat asthma to determine whether they possess pharmacological properties. Scientific investigations have shown that some of these herbs possess pharmacological and anti-inflammatory properties, and these may be useful in suppressing the characteristic exaggerated immune response in asthma [20-24]. Pepper and bayleaf have also been shown to exhibit anti-inflammatory properties [[25,26]27]. There is an imperative to commence scientific investigations on traditional West Indian medicinal plants to determine their therapeutic efficacy and safety.

The survey instrument specifically asked questions on the use of medicinal herbs in asthma and did not inquire about the use of herbs as customary teas or tonics. We therefore did not determine lifetime prevalence for the use of herbs in our patient sample, but we suppose that had this been included that there might have been a prevalence similar to those reported in the Jamaica [10,11] and Trinidad [13] surveys. The survey was also limited in that by electing to conduct the study at a public health facility we obviously had a bias towards patients at the lower rung of the socioeconomic ladder, with lower income and educational status. As a consequence, the results reflected patients from this demographic background. We may have expected a different outcome in asthmatic patients attending private institutions, where their characteristics would have been slightly different, as we noted that even in our sample the small number of persons with higher income and educational status tended to use more medicinal herbs for symptomatic relief.

We did not assess whether patients informed their attending physician at the clinic about their use of herbs or determined whether the knowledge or attitudes of these physicians regarding the use of herbs influenced the patients' decision to use herbs. The study was also limited in that we did not ascertain the out-of-pocket expense for herbal remedies by patients, although most stated that herbal medicines (which we supposed were processed, imported products) were more expensive than conventional medicines. We assumed that an additional expense would have only been incurred by those patients purchasing processed, imported herbs obtained from a herbalist, herbal shop or pharmacy (24.1%) and who actually consulted a herbalist (10.3%). We also reasoned that since all the other herbs used were inexpensive and available from either the backyard garden or supermarket (58.6%) that the cost to patients selecting these remedies was minimal.