Complementary and alternative medicine use in asthma: Who is using what?

Who has been surveyed and how?

The 17 identified studies were of poor to moderate quality only. A range of recruitment methods were utilized (see Table 2). The majority were conducted in Europe, the USA and Australia, with one study in Singapore, one in Turkey and one in India. Target populations included patients with asthma who were attending primary care physicians^10,16 specialist or hospital clinics,^{5,9,11–13,15,17,18,21} members of asthma organizations^7,14 and people with asthma identified after random digit dialling⁶ and population-based screening surveys.^8,19,20 The potential impact of these methods on the results is presented in Table 3. Questionnaire methods included face-to-face interviews^{5,8–11,13,15–18,20} telephone surveys,⁶ and self-completion surveys.^{7,12,14,19,21} The impact that this may have had on the results is presented in Table 4. The upper age limit for three of the five adult surveys was 50 years, and the paediatric surveys included caregivers of patients from 0 to 19 years.

What definitions of CAM were used in the surveys?

One of the most striking features of this review is the wide range of definitions or descriptors used for CAM in the questionnaires. For example, Ernst,^7,14 Janson et al.,⁸ Shenfield et al.¹⁸ and Mazur et al.¹⁵ used broad, open-ended terms such as ‘complementary therapies/medicines’ and ‘remedies other than drug treatment’ when enquiring about CAM use. In contrast, Blanc et al.,^5,6 Ng et al.,¹⁰ Andrews et al.,¹² Pachter et al.,¹⁶ Partridge et al.²⁰ and Singh et al.²¹ used narrower definitions of CAM by providing their respondents with a list of specific modalities to elicit responses about CAM use. Rivera et al.,¹¹ Braganza et al.,¹³ Orhan et al.¹⁷ and Reznik et al.¹⁹ included general questions about CAM use as well as providing a list of specific treatments. As highlighted by the Cochrane Collaboration’s definition of CAM,¹ what is considered to be CAM may vary between countries. In many Asian countries, such as China, an integrative approach to CAM use has been taken.²³ Therefore, traditional herbal remedies may be considered part of usual first line treatment, whereas these would be considered strictly CAM in the West.

What proportion of people with asthma use CAM?

As is evident from Table 2, there is large variability in the reported prevalence rates of use of CAM. Importantly, what constitutes ‘usage’ has not been defined in terms of frequency or consistency of use in any of the included studies. That is, one-off use cannot be differentiated from continuous use. In the paediatric surveys, 33%−89% of respondents were reported to use CAM. The prevalence of CAM use by adults ranged between 4–79%. The source of this variability is primarily explained by the differing methods employed. The reported prevalence of CAM use for adults was lower in studies with a sample size greater than 500 (Fig. 1), where the sample was chosen randomly rather than from a convenience sample (Fig. 2), where the quality score is higher (Fig. 3) and when the time frame considered was limited to the previous 12 months (Fig. 4). Although it appears that a lower prevalence of CAM use is found when adults are interviewed (Fig. 5) or where the survey response rate is greater than 75% (Fig. 6), caution must be exercised, as only one study utilized a self-completion questionnaire,⁷ and only one had a survey response rate greater than 75%.⁵ When an open-ended question was used, a slightly higher prevalence estimate was reported (Fig. 7). Therefore, on average, studies designed more rigorously estimate the prevalence of CAM use for asthma among adults as being between 20% and 30%.

For paediatrics, the reported prevalence of CAM use was lower in studies where the survey response rate was greater than 75% (Fig. 6), where the quality score is higher (Fig. 3) and when participants were responding to a questionnaire rather than being interviewed (Fig. 5). A lower prevalence of CAM use was observed in the only study of over 500 children (Fig. 1).¹⁴ Limiting the time frame to the previous 12 months (Fig. 4) did not appear to impact on the reported prevalence. When an open-ended question was used (Fig. 7), a slightly lower prevalence estimate was reported. No paediatric studies utilized a random sample (Fig. 2). Therefore, on average, studies designed more rigorously estimate the prevalence of CAM use for asthma among children as being between 50% and 60%.

Although the studies primarily asked about current, recent or ‘ever’ use of CAM, the surveys conducted by Ernst asked both the adult⁷ and paediatric¹⁴ subjects who were not current CAM users whether they would be likely to use CAM in the future. In the adult population, the majority (67%) thought that they may use CAM in the future compared with 27% who thought that they would not.⁷ Similarly, among caregivers for children with asthma, 71% thought that they would consider using CAM for their child in the future.¹⁴

What is being used?

The most commonly used forms of CAM reported by adults with asthma (Table 5) were: acupuncture, aromatherapy, breathing techniques, herbal products, homeopathy and yoga. Modalities used most frequently in the paediatric population were similar: breathing techniques, diet therapy, herbal medicine, homeopathy, massage, positive therapy, physical therapy, prayer, relaxation techniques, vitamins and minerals.

Do those using CAM perceive them to be useful?

Although objective evidence for the efficacy of CAM in asthma was not collected by the investigators of the included studies, several asked respondents about the perceived benefits of CAM. Patients interviewed by Lamb and Cantrill⁹ stated that they derived benefit from the use of relaxation/breathing exercises and yoga such as reduced symptoms. In Ernst’s⁷ survey of adults, the majority reported that their asthma symptoms had improved to ‘some’ or a ‘slight’ extent (response options ranged from ‘to a great extent’ to ‘not at all’) as a result of using CAM. The same investigator reported that the majority of caregivers of UK children with asthma¹⁴ felt that CAM had been ‘moderately useful’. Over half of the respondents recruited by Shenfield et al.¹⁸ who had used CAM were satisfied with them, although paradoxically only 12% stated that they believed them to be effective. Orhan et al.¹⁷ reported that 56% of their respondents perceived no improvement in their child’s asthma as a result of CAM use. In the study conducted by Pachter et al.¹⁶ respondents were asked whether they perceived the CAM modalities they had tried were effective. Affirmative responses depended on the modality and ranged from 0% (azogue = elemental mercury used in spiritual cleansing) to 75% (prayer). CAM was perceived to be as effective as conventional medicine by 59% of caregivers surveyed by Braganza et al.¹³ and by 59% adolescents surveyed by Reznik et al.¹⁹ A majority of those surveyed by Singh et al.²¹ did not report symptom relief from their use of CAM, but many believed that using their chosen CAM modality, in combination with strict dietary modification, would provide a cure in the long term.

Is there evidence to support the use of CAM for asthma?

Several comprehensive reviews summarizing the evidence for CAMs commonly used in asthma have been published previously, including two reviews of herbal interventions.^24,25 The reviewers of the herbal interventions conclude that although the effects of some products appear promising, it is difficult to interpret their efficacy because of the poor quality of the existing studies. Marks et al.²⁶ have compiled a review of common CAMs used in asthma including acupuncture, breathing techniques, chiropractic, homeopathy, meditation and yoga. Strong evidence for effectiveness for any of the modalities reviewed was not found.²⁶

How open are patients about their use of CAM?

In countries where an integrative approach to the use of CAM has not occurred, patient disclosure of CAM use is variable. Reasons for this include: reluctance to inform health-care practitioners for fear of ridicule, failure of health-care practitioners to directly ask patients, and the perception by patients that health-care practitioners would not be interested in knowing.^27,28 Mazur et al.¹⁵ were able to assess the difference between rates of CAM use spontaneously reported to medical practitioners during a normal consultation, and the rates subsequently elicited by direct questioning. At the child’s first visit, caregivers were questioned about their child’s current and past therapies. No parents volunteered that their child was using CAM. However, on direct questioning by the same physician at the subsequent appointment, 81% reported concomitant CAM usage. Shenfield et al.¹⁸ directly questioned caregivers on their level of disclosure regarding CAM to their child’s general practitioner/paediatrician. Only 48% had done so. A similar question was posed by Braganza et al.¹³ who reported that 82% had not informed their specialist, and by Reznik et al.¹⁹ who found that 46% had not informed their medical practitioner. The low level of disclosure about CAM by adolescents and caregivers is consistent with the findings in broader population surveys of CAM use.^2,3,29 In adults, Rivera et al.¹¹ were able to compare spontaneous reporting of CAM usage with direct questioning by comparing the results of their retrospective chart review with their subsequent interviews. No documentation of CAM use was found by chart review. In contrast, 42% of patients interviewed subsequently disclosed CAM use.

Given the low level of disclosure, can it be predicted who is most likely to be using CAM?

Several studies examined the associations between CAM utilization and demographic and disease characteristics. Among adults, Blanc et al.^5,6 and Ernst⁷ reported that there was a trend towards greater use of CAM among females compared with males. In contrast, Ng et al.¹⁰ found that in Singapore the use of CAM by males was more predominant, perhaps reflecting a cultural difference. More variation between studies was seen in associations with age and severity of asthma. In a randomly selected asthmatic population, Blanc et al. found that younger age was associated with a greater likelihood of self-treatment with caffeine⁶ but Ernst⁷ and Ng et al.¹⁰ were not able to detect an association between age and CAM use. Worse severity of disease has been associated with greater CAM use^7,10 and with use of coffee or black tea self-treatment in the clinic population study by Blanc et al.,⁵ but no such association was reported in the randomly selected asthmatic population subsequently studied by Blanc et al.⁶ A superior asthma knowledge score was found to be positively associated with CAM use by Ng et al.¹⁰

Factors predicting CAM use for children are also poorly characterized. Andrews et al.¹² found no association between the child’s age, duration or severity of asthma or presence of concurrent illness, with CAM use. Braganza et al.¹³ reported that CAM use was highest among those who were classified as having mild persistent or moderate persistent asthma. Shenfield et al.¹⁸ found that markers such as high-dose inhaled/oral steroids, persistent asthma, poor/very poor symptom control, frequency of bronchodilator side-effects and frequency of doctors’ visits were associated with CAM use. Orhan et al.¹⁷ and Reznik et al.¹⁹ found similar associations (regular asthma treatment, use of inhaled corticosteroids, asthma exacerbations, emergency admittances, frequency of asthma symptoms). Further, Orhan et al.¹⁷ and Reznik et al.¹⁹ found that intention to use CAM again in the future was associated with higher perceived efficacy of CAM.

From their interviews with adults, adolescents and caregivers, Partridge et al.²⁰ found that those expressing more concerns about conventional medications (feeling reliant on them, concerns about side-effects and long-term effects) were more likely to have used CAM.