Complementary Therapies in Clinical Practice 20 (2014) 302e310
Contents lists available at ScienceDirect
Complementary Therapies in Clinical Practice journal homepage: www.elsevier.com/locate/ctcp
Psychosocial factors that predict why people use complementary and alternative medicine and continue with its use: A population based study P. Thomson a, *, J. Jones b, M. Browne c, S.J. Leslie b, d a
School of Health Sciences, BG Bomont Building, University of Stirling, Stirling Campus, Stirling FK9 4LA, Scotland, UK School of Health Sciences, Centre for Health Science, University of Stirling, Highland Campus, Old Perth Road, Inverness, IV2 3JH, Scotland, UK School of Human, Health and Social Sciences, Central Queensland University, Bundaberg Campus, University Drive, Branyan QLD 4670, Australia d Cardiac Unit, Raigmore Hospital, Inverness IV32 3UJ, Scotland, UK b c
a b s t r a c t Keywords: Complementary and alternative medicine (CAM) Perceived control Cognitive style Spirituality Openness
Studies have explored the predictors of CAM use but fewer data explain the psychosocial factors associated with this and why people continue with CAM. Aims: To examine the psychosocial factors that predict CAM use; to explore the predictors of continuing with CAM. Design: A cross sectional survey. Methods: 1256 adults were interviewed as part of 2012 Queensland Social Survey. We included questions about CAM, perceived control, cognitive style, spirituality and openness. Relationships were explored using bivariate and multiple logistic regression. Results: 79% of people had used CAM in the last 12 months. Socio-demographics, health behaviours, spirituality, openness and prescribing sources were the strongest predictors of CAM use. General health, chronic illness and prescribing sources predicted continued CAM use. Conclusion: There was high CAM use in Queensland, Australia. Personal characteristics and psychosocial factors need to be considered as part of the individual's holistic assessment and on-going care. © 2014 Published by Elsevier Ltd.
1. Introduction The percentage of the general population that use CAM is increasing [1e5], although rates of CAM use reported in the literature vary widely from 10 to 52% [4e10]. There is a lack of consensual definition of what constitutes a CAM practice. CAM encompasses a wide variety of health-related philosophical approaches to disease [11]. Some CAM modalities are practitioner delivered (e.g. acupuncture, homeopathy, reflexology, massage) [12] and others involve self care practices (e.g. homeopathic remedies, herbal medicines, vitamins). Studies have shown that being female [1,3,4,6,13e18]; middle aged [19,20]; employed [3,6]; having a higher household income [1,3,5e8,15]; more education [1,6,14,15,19]; particular ethnicity [19,20]; and geographic location [1,3,4,15,21] are associated with increased CAM use. Other factors include positive health behaviours such as not smoking, eating a healthy diet and being * Corresponding author. Tel.: þ44(0) 1786 466396. E-mail address: [email protected] (P. Thomson). http://dx.doi.org/10.1016/j.ctcp.2014.09.004 1744-3881/© 2014 Published by Elsevier Ltd.
physically active; and preventative strategies such as stress management and avoiding excessive drinking are associated with increased CAM use [4,22]. Also associated with increase CAM use are poorer health status [7,19e21,23], and chronic health problems [5,15,24e27] and greater perceived control over health [15,28]. Hildreth et al. (2007) and Astin (1998) found that some people view their health as a reflection of religious and spiritual beliefs [19,29]. Religiousness has been defined as a form of subscription to institutionalised doctrines or beliefs [30], whereas spirituality is described as adhering to the notion that there is an unknown ‘essence of the human person’ [31], or a belief in a ‘transcendental force’ [32]. Hildreth et al. (2007) found that selfrated spirituality but not religiosity was associated with greater likelihood of being a CAM user [29]. Ellison et al. (2012) identify that individuals high in spirituality were more likely to use CAM, but religiosity was negatively related to CAM use [33]. Survey findings from the US indicate that after controlling for established predictors of CAM use, such as education and personality, both spirituality and religiousness were associated in unique ways with using CAM [33]. It is also considered that post-modern lifestyles
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
and values, such as feminism and vegetarianism, as well as spirituality positively predict using CAM. This suggests a philosophical mind-set which values holistic balance and personal growth which is congruent with CAM use [34]. The personality dimension of openness to new experiences is also associated with an increased use of CAM [35,36]. Openness is one feature of the well-established Five Factor Model [37], which describes the extent to which an individual is open to new ideas, approaches and experiences [37,38]. This trait is a distinct dimension [29], one not usually associated with increased conventional care use. Openness, spirituality and mood attention are also associated with people decisions to use CAM and to use most types of CAM [36]. Fewer studies have explored the reasons for continuing with CAM use and previous data have been limited as it focuses primarily on practitioner delivered CAM. Available literature indicates that interpersonal (interactions with practitioners); physical (sensations such as touch or pain during treatment); affective (empowerment); and cognitive (beliefs about treatment) [35,37] factors are influential in continued CAM use. Sirois and Gick (2012) have found committed CAM users to be more motivated by medical need, such as chronic pain [35]. Health-aware behaviours also predict continued use of CAM since greater experience may reinforce an awareness of and the practice of healthy behaviours [35]. Still less is known about the potential influence of ‘cognitive style’ and its association with CAM use, and peoples' reasons for continuing with CAM from the perspective of consumers. Cognitive style describes one's preference towards fast, heuristic thinking (intuitive) versus slow, deliberate and logical thinking (analytical) [39]. Such information may help minimise the risks associated with impulsive decision making and help promote shared decision making with respect to CAM and on-going care [40,41]. This current study aimed to examine the psychosocial factors that predict CAM use; and to explore the predictors of continuing with CAM in the adult population of Queensland, Australia. 2. Methods 2.1. Design and sample This cross-sectional study employed data from the 2012 Queensland Social Survey (QSS) [42]. In a linked paper, we explored the factors associated with intention to try CAM before conventional medicine and the predictors of initially seeking CAM in the same population [43]. A two-stage stratified sampling strategy was used to randomly select households and individuals. The sample was derived from the commercially available Electronic White Pages using a computer program. Within each contacted household, one eligible person was selected for the interview based on age (18 years or older), sex and availability. Survey estimates of sampling error for the total sample of 1256 showed this was accurate within ±2.8 percentage points at a 95% CI. The PRL used the index of dissimilarity for age distributions to provide a measure of sample representativeness [44]. The most recent Australian Bureau of Statistics [45] census data was used for comparison with the 2012 QSS sample. 2.2. Measures The 2012 QSS contained questions on socio-demographics, health status, personal health behaviours and chronic health problems. We added questions/statements about CAM (Appendix 1). 2.2.1. Socio-demographics Age was originally categorised as: 18e24 years, 25e34 years, 35e44 years, 45e54 years, 55e64 years and 65 or older (later re-
303
coded as under 55 and 55þ years to allow for additional statistical analysis). Marital status was classified as married, de facto (cohabitating), separated/divorced, widowed and single (recoded as married/partnered and unmarried/un-partnered). Employment was defined as being in paid employment in the previous week or not, categorised as employed (full-time), employed (part-time/casual), unemployed, retired/pensioner, student and home duties (recoded as employed and not employed/no response). Years of education, place of residence, household income and religion were identified; and country of birth was recorded as Australia or other. 2.2.2. Perceived health status General health was assessed using one item from the Healthy Days Core Module [46]. Physical health and mental health and combined physical and mental health were assessed by identifying the number of days the individual had not been good in the last 30 days. Respondents were also asked to identify if they had a chronic health problem. 2.2.3. Health behaviours Smoking status, height and weight, daily fruit and vegetable consumption, fast food consumption were identified by self report. BMI was calculated from height (in centimetres) and weight (in kilograms). Alcohol consumption, considering all types of alcoholic beverages, was identified according to how many times during the past 30 days the respondents had 6 or more drinks on an occasion [46]. Self-reported leisure-time physical activity was identified in accordance with the Active Australia physical activity scale [47], recoded as sufficient (non-sedentary) if the respondents had spent any time in physical activities i.e. walking, moderate or vigorous intensity activities in the week prior to the survey, or insufficient (sedentary) if they had spent no time in these activities. 2.2.4. Complementary and alternative therapies The CAM questions/statements identified were derived from previously validated questionnaires [48] and from studies that have distinguished between people who have used CAM from those who have not [4,6,49], the reasons why people use CAM [5,13,19e21,26,50], and continue with its use [15], to include such factors as perceived control over health [50], cognitive style [39], spirituality and religion [29,33], and openness to new experiences [35,36]. In the survey, CAM is defined as both practitioner delivered therapies and self-care practices provided alongside or instead of conventional medicine. 2.3. Data collection The use of ethical research protocols and trained interviewers helped ensure the quality of data collected. Pilot testing by trained interviewers on 56 randomly selected households allowed modification to the final questionnaire. Interviewing for the main survey (second round) began in October 2012 and was completed by December 2012. Approval for the study was obtained from the Human Ethics Research Panel at CQUniversity before administration to the general public (Project: H10/06-121, QSS 2012). All subjects gave informed consent to the research. 2.4. Statistical analysis Data analyses were performed using Predictive Analytics Software (PASW) statistics version 19. P < 0.05 was taken to indicate statistical significance. Descriptive statistics were computed for the CAM and cognitive style questions/related statements, and correlations examined using Spearman's rho. Bivariate relationships
304
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
were examined for the two dichotomous outcomes: 1) CAM use (yes/no); 2) reasons for continuing with CAM (medical vs nonmedical reasons) and the independent variables e sociodemographics, health status, health behaviours, perceived control over health, cognitive style, openness to new experiences, spirituality, religion, prescribing sources and chronic health problem, using logistic regression. Multiple logistic regression (backward stepwise, likelihood ratio) was conducted using HosmereLemeshow goodness-of-fit tests. Only the independent variables statistically significant in the bivariate analysis were included in the multiple regression models. All variables were initially entered into the model then removed stepwise (with the possibility of inclusion in the model of p < 0.05 and for removal p > 0.10) to identify the independent predictors.
3.3. Complementary and alternative medicine The types of CAM used and those who prescribed or recommended them are presented in Table 3. Also shown in Table 3 are the results for correlations between CAM practices and cognitive style. Results revealed significant associations for herbs, prayer, and homeopathy. This suggests that utilisation of some forms of complementary practices are associated with a preference for evaluating treatment alternatives from an intuitive, rather than analytical, cognitive stance. Table 2 presents the main reasons given for continuing using CAM. Additional sources of information on CAM were identified as pharmacists, naturopaths, podiatrists and other sources such as television adverts, the radio or internet. 4. Discussion
3. Results There was a 40.3% response rate. Socio-demographic data and information on health behaviours, general health, personal control over health, cognitive style, spirituality, openness, prescribing sources and chronic health problems are presented in Table 1. 3.1. Factors associated with CAM use Results revealed that amongst the 1256 respondents, 79% of people had used at least one form of CAM (practitioner delivered or self care practices) in the last 12 months. Table 1 presents the unadjusted and adjusted odds ratios (with 95% confidence intervals) for CAM use (yes/no), associated with the independent variables. In the adjusted model, the independent predictors of CAM use were gender, education, smoking, spirituality, openness and prescribing sources. There was a decreased likelihood of males, less educated individuals and smokers using CAM, compared to females, more educated individuals and non-smokers. In relation to spirituality, the respondents who agreed (or neither agreed or disagreed) the most important knowledge comes from spiritual experience were over twice as likely to use CAM, compared to those who disagreed. Those who agreed they were open to new experiences were also more likely to use CAM, compared to those who disagreed. Respondents prescribed or recommended CAM by a medical practitioner were less likely to use CAM, compared to non-medical prescribers. In the unadjusted model, there was a suggestion that individuals under the age of 55 years were more inclined to use CAM, compared to those 55 years and over (Table 1). 3.2. Factors associated with the reasons for continuing with CAM Table 1 presents the unadjusted and adjusted odds ratios (with 95% confidence intervals) for continuing with CAM (medical vs non-medical reasons), associated with the independent variables. We limited our analysis to the highest percentage response categories i.e. to add to conventional medical treatment of a condition (medical reasons); and general well-being or to treat a nonmedical condition (non-medical reasons) (Table 2). In the adjusted model, the independent predictors for continuing with CAM were general health, prescribing sources and having a chronic health problem (Table 1). People with excellent or very good health were over three times more likely to use CAM for nonmedical reasons i.e. for health and well-being, compared to those with good, fair or poor health. Those prescribed or recommended CAM by a medical practitioner were less likely to continue using CAM (for non-medical reasons). There was a decreased likelihood of people continuing with CAM for non-medical reasons if there was a chronic health problem.
This study examined the psychosocial factors that predict CAM use and the predictors of continuing with CAM in the adult population of Queensland, Australia. A high proportion of people used CAM (practitioner delivered or self care practices) in the last 12 months then previously. This reflects a trend of increasing CAM use which is consistent with other studies' and review findings from Australia and other countries [1e3,51]. Respondents had used at least one of the therapies identified in the House of Lords Report on CAM [11], including acupuncture, chiropractic, herbal medicine and osteopathy which confirms previous findings reported by Thomas & Coleman [8]. Our most significant study finding was that spirituality independently predicted increased CAM use. This is consistent with previous research [19,29] that suggest spirituality positively predicts CAM use, which also concurs with post-modern lifestyles and values [34]. This finding contributes to a small but growing literature on the links between CAM use and personal values, spirituality and personality [19,29,33,34,52,53]. Religion was not associated with CAM use and this is consistent with previous findings that self rated religiosity and spirituality have differential associations with CAM use [19,29]. Another significant finding was that openness to new experiences significantly predicted an increased likelihood of using CAM. Our results are consistent with previous research that found both openness to experience and spirituality are strong predictors of willingness to use CAM [36], and to use particular types of CAM [52]. People with this personality trait (i.e. openness) are considered more open to new ideas, approaches and experiences [37]. They also tend to use less conventional care and are less adherent to conventional belief systems [54]. Our findings for personal control over health are contrary to previous research which suggest an association between control and CAM use [15,28]. People with greater perceived control are thought more inclined to use CAM although there are some mixed findings from research with respect to this [29]. We found no increased or decreased likelihood of using CAM by cognitive style, which describes one's preference towards a particular type of thinking [39]. No previous studies were found for direct comparison. Our negative finding indicates this cognitive trait is not associated with the behavioural measure of using CAM. More complex cognitive processes may be at play here as part of the logic of selecting CAM. Also, the three-item Cognitive Reflective Test (CRT) [39] may not be sufficiently sensitive to show rational thought processes as a predictor of CAM use. Similar to previous studies we found that being female [1,3,4,6,13e18] was a significant independent predictors of CAM use. In our study, fewer years of education was a significant predictor of a decreased likelihood of using CAM, which is consistent with previous research that identify greater CAM use in more educated individuals [1,6,14,15,19]. There was no increased or
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
305
Table 1 Regression analysis showing odds ratios (ORs), 95% confidence intervals (CI) P values for socio-demographics, general health, beliefs and experiences and: 1) use of complementary and alternative medicine (CAM) and 2) reasons for continuing with CAM. CAM use (n ¼ 1256)
Reasons for continuing with CAM (n ¼ 408) Adjusted ORa
Unadjusted OR
Adjusted ORa
Unadjusted OR P 0.305
e
P e
0.99 (0.98e1.01) 1.00 ()
0.911
e
e
23.0 77.0
1.02 (0.99e1.06) 1.00 ()
0.163
e
e
0.001 0.131 0.172
27.5 18.9 12.5 41.2
0.98 0.91 0.95 1.00
(0.60e1.61) (0.53e1.58) (0.50e1.81) ()
0.965 0.744 0.881
e
e
e
e
55.6 44.4
1.39 (0.93e2.08) 1.00 ()
0.106
e
e
0.195 0.051 0.358
e
e
50.0 10.5 16.2 23.3
1.07 0.54 0.59 1.00
(0.64e1.77) (0.26e1.10) (0.31e1.12) ()
0.803 0.090 0.108
e
e
0.75 (0.51e1.11) 1.00 ()
0.148
e
e
83.6 16.4
0.93 (0.54e1.59) 1.00 ()
0.795
e
e
59.1 40.9
1.28 (0.93e1.61) 1.00 ()
0.144
e
e
62.0 38.0
0.80 (0.53e1.21) 1.00 ()
0.293
e
e
Smoking Smoker Non-smoker
11.5 88.5
0.66 (0.44e0.97) 1.00 ()
0.003
0.63 (0.42e0.96) 1.00 ()
0.030
10.8 89.2
1.12 (0.58e2.14) 1.00 ()
0.734
e
e
Body mass index (BMI) 18.4e24.9 >25.0
36.0 64.0
1.38 (0.98e1.93) 1.00 ()
0.622
e
e
35.3 64.7
1.67 (1.01e2.76) 1.00 ()
0.054
e
e
Unhealthy diet (McDonalds etc) Never Once to more than times
61.7 38.3
0.77 (0.57e1.02) 1.00 ()
0.072
e
e
62.5 37.5
0.87 (0.57e1.31) 1.00 ()
0.490
e
e
Fruit servings None (per day) Less than 2 2þ per day
13.4 29.9 56.7
0.72 (0.48e1.08) 0.77 (0.57e1.05) 1.00 ()
0.111 0.101
e
e
14.0 30.4 55.6
1.03 (0.56e1.86) 1.05 (0.67e1.66) 1.00 ()
0.933 0.811
e
e
Physical activity Sufficient (non-sedentary) Insufficient (sedentary)
47.7 52.3
1,65 (1.14e2.39) 1.00 ()
0.008
0.91 (0.68e1.22) 1.00 ()
0.545
46.3 53.7
1.62 (0.88e3.01) 1.00 ()
0.122
e
e
General health status Excellent Very good Good Fair Poor
15.4 35.0 32.4 12.3 4.9
1.30 0.97 1.18 1.34 1.00
(0.65e2.63) (0.52e1.84) (0.62e2.25) (0.65e2.76) ()
0.455 0.941 0.611 0.432
e
e
15.9 35.0 32.4 12.0 4.7
6.12 5.21 2.76 1.62 1.00
0.001 0.002 0.052 0.396
3.47 3.35 1.90 1.48 1.00
Beliefs (personal control, if sick) Disagree (strongly/mod/slightly) Agree (strongly/mod/slightly) Don't know/did not use
18.2 79.1 2.7
0.75 (0.31e1.81) 0.99 (0.43e2.32) 1.00 ()
0.522 0.995
e
e
17.4 80.6 2.0
1.29 (0.29e5.57) 1.67 (0.41e6.82) 1.0 ()
0.733 0.472
e
e
Cognitive style Total score (mean)
e
0.97 (0.94e1.01)
0.117
e
e
e
1.05 (0.97e1.15)
0.197
e
e
Knowledge (spiritual experience) Disagree (strongly/mod/little) Neither agree nor disagree/DK Agree (little/mod/strongly)
57.2 13.0 29.9
1.00 () 2.08 (1.22e3.58) 2.74 (1.52e4.94)
1.00 () 2.17 (1.36e3.44) 2.83 (1.37e5.82)
1.30 (0.81e2.08) 1.14 (0.67e1.99) 1.0 ()
0.275 0.626
e
0.001 0.005
55.6 16.9 27.5
e
0.008 0.001
Openness (to new experiences) Disagree (strongly/mod/little)
6.9
1.00 ()
7.6
1.00 ()
Gender Male Female
% 50.3 49.7
0.44 (0.33e0.58) 1.00 ()
P
Contents lists available at ScienceDirect
Complementary Therapies in Clinical Practice journal homepage: www.elsevier.com/locate/ctcp
Psychosocial factors that predict why people use complementary and alternative medicine and continue with its use: A population based study P. Thomson a, *, J. Jones b, M. Browne c, S.J. Leslie b, d a
School of Health Sciences, BG Bomont Building, University of Stirling, Stirling Campus, Stirling FK9 4LA, Scotland, UK School of Health Sciences, Centre for Health Science, University of Stirling, Highland Campus, Old Perth Road, Inverness, IV2 3JH, Scotland, UK School of Human, Health and Social Sciences, Central Queensland University, Bundaberg Campus, University Drive, Branyan QLD 4670, Australia d Cardiac Unit, Raigmore Hospital, Inverness IV32 3UJ, Scotland, UK b c
a b s t r a c t Keywords: Complementary and alternative medicine (CAM) Perceived control Cognitive style Spirituality Openness
Studies have explored the predictors of CAM use but fewer data explain the psychosocial factors associated with this and why people continue with CAM. Aims: To examine the psychosocial factors that predict CAM use; to explore the predictors of continuing with CAM. Design: A cross sectional survey. Methods: 1256 adults were interviewed as part of 2012 Queensland Social Survey. We included questions about CAM, perceived control, cognitive style, spirituality and openness. Relationships were explored using bivariate and multiple logistic regression. Results: 79% of people had used CAM in the last 12 months. Socio-demographics, health behaviours, spirituality, openness and prescribing sources were the strongest predictors of CAM use. General health, chronic illness and prescribing sources predicted continued CAM use. Conclusion: There was high CAM use in Queensland, Australia. Personal characteristics and psychosocial factors need to be considered as part of the individual's holistic assessment and on-going care. © 2014 Published by Elsevier Ltd.
1. Introduction The percentage of the general population that use CAM is increasing [1e5], although rates of CAM use reported in the literature vary widely from 10 to 52% [4e10]. There is a lack of consensual definition of what constitutes a CAM practice. CAM encompasses a wide variety of health-related philosophical approaches to disease [11]. Some CAM modalities are practitioner delivered (e.g. acupuncture, homeopathy, reflexology, massage) [12] and others involve self care practices (e.g. homeopathic remedies, herbal medicines, vitamins). Studies have shown that being female [1,3,4,6,13e18]; middle aged [19,20]; employed [3,6]; having a higher household income [1,3,5e8,15]; more education [1,6,14,15,19]; particular ethnicity [19,20]; and geographic location [1,3,4,15,21] are associated with increased CAM use. Other factors include positive health behaviours such as not smoking, eating a healthy diet and being * Corresponding author. Tel.: þ44(0) 1786 466396. E-mail address: [email protected] (P. Thomson). http://dx.doi.org/10.1016/j.ctcp.2014.09.004 1744-3881/© 2014 Published by Elsevier Ltd.
physically active; and preventative strategies such as stress management and avoiding excessive drinking are associated with increased CAM use [4,22]. Also associated with increase CAM use are poorer health status [7,19e21,23], and chronic health problems [5,15,24e27] and greater perceived control over health [15,28]. Hildreth et al. (2007) and Astin (1998) found that some people view their health as a reflection of religious and spiritual beliefs [19,29]. Religiousness has been defined as a form of subscription to institutionalised doctrines or beliefs [30], whereas spirituality is described as adhering to the notion that there is an unknown ‘essence of the human person’ [31], or a belief in a ‘transcendental force’ [32]. Hildreth et al. (2007) found that selfrated spirituality but not religiosity was associated with greater likelihood of being a CAM user [29]. Ellison et al. (2012) identify that individuals high in spirituality were more likely to use CAM, but religiosity was negatively related to CAM use [33]. Survey findings from the US indicate that after controlling for established predictors of CAM use, such as education and personality, both spirituality and religiousness were associated in unique ways with using CAM [33]. It is also considered that post-modern lifestyles
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
and values, such as feminism and vegetarianism, as well as spirituality positively predict using CAM. This suggests a philosophical mind-set which values holistic balance and personal growth which is congruent with CAM use [34]. The personality dimension of openness to new experiences is also associated with an increased use of CAM [35,36]. Openness is one feature of the well-established Five Factor Model [37], which describes the extent to which an individual is open to new ideas, approaches and experiences [37,38]. This trait is a distinct dimension [29], one not usually associated with increased conventional care use. Openness, spirituality and mood attention are also associated with people decisions to use CAM and to use most types of CAM [36]. Fewer studies have explored the reasons for continuing with CAM use and previous data have been limited as it focuses primarily on practitioner delivered CAM. Available literature indicates that interpersonal (interactions with practitioners); physical (sensations such as touch or pain during treatment); affective (empowerment); and cognitive (beliefs about treatment) [35,37] factors are influential in continued CAM use. Sirois and Gick (2012) have found committed CAM users to be more motivated by medical need, such as chronic pain [35]. Health-aware behaviours also predict continued use of CAM since greater experience may reinforce an awareness of and the practice of healthy behaviours [35]. Still less is known about the potential influence of ‘cognitive style’ and its association with CAM use, and peoples' reasons for continuing with CAM from the perspective of consumers. Cognitive style describes one's preference towards fast, heuristic thinking (intuitive) versus slow, deliberate and logical thinking (analytical) [39]. Such information may help minimise the risks associated with impulsive decision making and help promote shared decision making with respect to CAM and on-going care [40,41]. This current study aimed to examine the psychosocial factors that predict CAM use; and to explore the predictors of continuing with CAM in the adult population of Queensland, Australia. 2. Methods 2.1. Design and sample This cross-sectional study employed data from the 2012 Queensland Social Survey (QSS) [42]. In a linked paper, we explored the factors associated with intention to try CAM before conventional medicine and the predictors of initially seeking CAM in the same population [43]. A two-stage stratified sampling strategy was used to randomly select households and individuals. The sample was derived from the commercially available Electronic White Pages using a computer program. Within each contacted household, one eligible person was selected for the interview based on age (18 years or older), sex and availability. Survey estimates of sampling error for the total sample of 1256 showed this was accurate within ±2.8 percentage points at a 95% CI. The PRL used the index of dissimilarity for age distributions to provide a measure of sample representativeness [44]. The most recent Australian Bureau of Statistics [45] census data was used for comparison with the 2012 QSS sample. 2.2. Measures The 2012 QSS contained questions on socio-demographics, health status, personal health behaviours and chronic health problems. We added questions/statements about CAM (Appendix 1). 2.2.1. Socio-demographics Age was originally categorised as: 18e24 years, 25e34 years, 35e44 years, 45e54 years, 55e64 years and 65 or older (later re-
303
coded as under 55 and 55þ years to allow for additional statistical analysis). Marital status was classified as married, de facto (cohabitating), separated/divorced, widowed and single (recoded as married/partnered and unmarried/un-partnered). Employment was defined as being in paid employment in the previous week or not, categorised as employed (full-time), employed (part-time/casual), unemployed, retired/pensioner, student and home duties (recoded as employed and not employed/no response). Years of education, place of residence, household income and religion were identified; and country of birth was recorded as Australia or other. 2.2.2. Perceived health status General health was assessed using one item from the Healthy Days Core Module [46]. Physical health and mental health and combined physical and mental health were assessed by identifying the number of days the individual had not been good in the last 30 days. Respondents were also asked to identify if they had a chronic health problem. 2.2.3. Health behaviours Smoking status, height and weight, daily fruit and vegetable consumption, fast food consumption were identified by self report. BMI was calculated from height (in centimetres) and weight (in kilograms). Alcohol consumption, considering all types of alcoholic beverages, was identified according to how many times during the past 30 days the respondents had 6 or more drinks on an occasion [46]. Self-reported leisure-time physical activity was identified in accordance with the Active Australia physical activity scale [47], recoded as sufficient (non-sedentary) if the respondents had spent any time in physical activities i.e. walking, moderate or vigorous intensity activities in the week prior to the survey, or insufficient (sedentary) if they had spent no time in these activities. 2.2.4. Complementary and alternative therapies The CAM questions/statements identified were derived from previously validated questionnaires [48] and from studies that have distinguished between people who have used CAM from those who have not [4,6,49], the reasons why people use CAM [5,13,19e21,26,50], and continue with its use [15], to include such factors as perceived control over health [50], cognitive style [39], spirituality and religion [29,33], and openness to new experiences [35,36]. In the survey, CAM is defined as both practitioner delivered therapies and self-care practices provided alongside or instead of conventional medicine. 2.3. Data collection The use of ethical research protocols and trained interviewers helped ensure the quality of data collected. Pilot testing by trained interviewers on 56 randomly selected households allowed modification to the final questionnaire. Interviewing for the main survey (second round) began in October 2012 and was completed by December 2012. Approval for the study was obtained from the Human Ethics Research Panel at CQUniversity before administration to the general public (Project: H10/06-121, QSS 2012). All subjects gave informed consent to the research. 2.4. Statistical analysis Data analyses were performed using Predictive Analytics Software (PASW) statistics version 19. P < 0.05 was taken to indicate statistical significance. Descriptive statistics were computed for the CAM and cognitive style questions/related statements, and correlations examined using Spearman's rho. Bivariate relationships
304
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
were examined for the two dichotomous outcomes: 1) CAM use (yes/no); 2) reasons for continuing with CAM (medical vs nonmedical reasons) and the independent variables e sociodemographics, health status, health behaviours, perceived control over health, cognitive style, openness to new experiences, spirituality, religion, prescribing sources and chronic health problem, using logistic regression. Multiple logistic regression (backward stepwise, likelihood ratio) was conducted using HosmereLemeshow goodness-of-fit tests. Only the independent variables statistically significant in the bivariate analysis were included in the multiple regression models. All variables were initially entered into the model then removed stepwise (with the possibility of inclusion in the model of p < 0.05 and for removal p > 0.10) to identify the independent predictors.
3.3. Complementary and alternative medicine The types of CAM used and those who prescribed or recommended them are presented in Table 3. Also shown in Table 3 are the results for correlations between CAM practices and cognitive style. Results revealed significant associations for herbs, prayer, and homeopathy. This suggests that utilisation of some forms of complementary practices are associated with a preference for evaluating treatment alternatives from an intuitive, rather than analytical, cognitive stance. Table 2 presents the main reasons given for continuing using CAM. Additional sources of information on CAM were identified as pharmacists, naturopaths, podiatrists and other sources such as television adverts, the radio or internet. 4. Discussion
3. Results There was a 40.3% response rate. Socio-demographic data and information on health behaviours, general health, personal control over health, cognitive style, spirituality, openness, prescribing sources and chronic health problems are presented in Table 1. 3.1. Factors associated with CAM use Results revealed that amongst the 1256 respondents, 79% of people had used at least one form of CAM (practitioner delivered or self care practices) in the last 12 months. Table 1 presents the unadjusted and adjusted odds ratios (with 95% confidence intervals) for CAM use (yes/no), associated with the independent variables. In the adjusted model, the independent predictors of CAM use were gender, education, smoking, spirituality, openness and prescribing sources. There was a decreased likelihood of males, less educated individuals and smokers using CAM, compared to females, more educated individuals and non-smokers. In relation to spirituality, the respondents who agreed (or neither agreed or disagreed) the most important knowledge comes from spiritual experience were over twice as likely to use CAM, compared to those who disagreed. Those who agreed they were open to new experiences were also more likely to use CAM, compared to those who disagreed. Respondents prescribed or recommended CAM by a medical practitioner were less likely to use CAM, compared to non-medical prescribers. In the unadjusted model, there was a suggestion that individuals under the age of 55 years were more inclined to use CAM, compared to those 55 years and over (Table 1). 3.2. Factors associated with the reasons for continuing with CAM Table 1 presents the unadjusted and adjusted odds ratios (with 95% confidence intervals) for continuing with CAM (medical vs non-medical reasons), associated with the independent variables. We limited our analysis to the highest percentage response categories i.e. to add to conventional medical treatment of a condition (medical reasons); and general well-being or to treat a nonmedical condition (non-medical reasons) (Table 2). In the adjusted model, the independent predictors for continuing with CAM were general health, prescribing sources and having a chronic health problem (Table 1). People with excellent or very good health were over three times more likely to use CAM for nonmedical reasons i.e. for health and well-being, compared to those with good, fair or poor health. Those prescribed or recommended CAM by a medical practitioner were less likely to continue using CAM (for non-medical reasons). There was a decreased likelihood of people continuing with CAM for non-medical reasons if there was a chronic health problem.
This study examined the psychosocial factors that predict CAM use and the predictors of continuing with CAM in the adult population of Queensland, Australia. A high proportion of people used CAM (practitioner delivered or self care practices) in the last 12 months then previously. This reflects a trend of increasing CAM use which is consistent with other studies' and review findings from Australia and other countries [1e3,51]. Respondents had used at least one of the therapies identified in the House of Lords Report on CAM [11], including acupuncture, chiropractic, herbal medicine and osteopathy which confirms previous findings reported by Thomas & Coleman [8]. Our most significant study finding was that spirituality independently predicted increased CAM use. This is consistent with previous research [19,29] that suggest spirituality positively predicts CAM use, which also concurs with post-modern lifestyles and values [34]. This finding contributes to a small but growing literature on the links between CAM use and personal values, spirituality and personality [19,29,33,34,52,53]. Religion was not associated with CAM use and this is consistent with previous findings that self rated religiosity and spirituality have differential associations with CAM use [19,29]. Another significant finding was that openness to new experiences significantly predicted an increased likelihood of using CAM. Our results are consistent with previous research that found both openness to experience and spirituality are strong predictors of willingness to use CAM [36], and to use particular types of CAM [52]. People with this personality trait (i.e. openness) are considered more open to new ideas, approaches and experiences [37]. They also tend to use less conventional care and are less adherent to conventional belief systems [54]. Our findings for personal control over health are contrary to previous research which suggest an association between control and CAM use [15,28]. People with greater perceived control are thought more inclined to use CAM although there are some mixed findings from research with respect to this [29]. We found no increased or decreased likelihood of using CAM by cognitive style, which describes one's preference towards a particular type of thinking [39]. No previous studies were found for direct comparison. Our negative finding indicates this cognitive trait is not associated with the behavioural measure of using CAM. More complex cognitive processes may be at play here as part of the logic of selecting CAM. Also, the three-item Cognitive Reflective Test (CRT) [39] may not be sufficiently sensitive to show rational thought processes as a predictor of CAM use. Similar to previous studies we found that being female [1,3,4,6,13e18] was a significant independent predictors of CAM use. In our study, fewer years of education was a significant predictor of a decreased likelihood of using CAM, which is consistent with previous research that identify greater CAM use in more educated individuals [1,6,14,15,19]. There was no increased or
P. Thomson et al. / Complementary Therapies in Clinical Practice 20 (2014) 302e310
305
Table 1 Regression analysis showing odds ratios (ORs), 95% confidence intervals (CI) P values for socio-demographics, general health, beliefs and experiences and: 1) use of complementary and alternative medicine (CAM) and 2) reasons for continuing with CAM. CAM use (n ¼ 1256)
Reasons for continuing with CAM (n ¼ 408) Adjusted ORa
Unadjusted OR
Adjusted ORa
Unadjusted OR P 0.305
e
P e
0.99 (0.98e1.01) 1.00 ()
0.911
e
e
23.0 77.0
1.02 (0.99e1.06) 1.00 ()
0.163
e
e
0.001 0.131 0.172
27.5 18.9 12.5 41.2
0.98 0.91 0.95 1.00
(0.60e1.61) (0.53e1.58) (0.50e1.81) ()
0.965 0.744 0.881
e
e
e
e
55.6 44.4
1.39 (0.93e2.08) 1.00 ()
0.106
e
e
0.195 0.051 0.358
e
e
50.0 10.5 16.2 23.3
1.07 0.54 0.59 1.00
(0.64e1.77) (0.26e1.10) (0.31e1.12) ()
0.803 0.090 0.108
e
e
0.75 (0.51e1.11) 1.00 ()
0.148
e
e
83.6 16.4
0.93 (0.54e1.59) 1.00 ()
0.795
e
e
59.1 40.9
1.28 (0.93e1.61) 1.00 ()
0.144
e
e
62.0 38.0
0.80 (0.53e1.21) 1.00 ()
0.293
e
e
Smoking Smoker Non-smoker
11.5 88.5
0.66 (0.44e0.97) 1.00 ()
0.003
0.63 (0.42e0.96) 1.00 ()
0.030
10.8 89.2
1.12 (0.58e2.14) 1.00 ()
0.734
e
e
Body mass index (BMI) 18.4e24.9 >25.0
36.0 64.0
1.38 (0.98e1.93) 1.00 ()
0.622
e
e
35.3 64.7
1.67 (1.01e2.76) 1.00 ()
0.054
e
e
Unhealthy diet (McDonalds etc) Never Once to more than times
61.7 38.3
0.77 (0.57e1.02) 1.00 ()
0.072
e
e
62.5 37.5
0.87 (0.57e1.31) 1.00 ()
0.490
e
e
Fruit servings None (per day) Less than 2 2þ per day
13.4 29.9 56.7
0.72 (0.48e1.08) 0.77 (0.57e1.05) 1.00 ()
0.111 0.101
e
e
14.0 30.4 55.6
1.03 (0.56e1.86) 1.05 (0.67e1.66) 1.00 ()
0.933 0.811
e
e
Physical activity Sufficient (non-sedentary) Insufficient (sedentary)
47.7 52.3
1,65 (1.14e2.39) 1.00 ()
0.008
0.91 (0.68e1.22) 1.00 ()
0.545
46.3 53.7
1.62 (0.88e3.01) 1.00 ()
0.122
e
e
General health status Excellent Very good Good Fair Poor
15.4 35.0 32.4 12.3 4.9
1.30 0.97 1.18 1.34 1.00
(0.65e2.63) (0.52e1.84) (0.62e2.25) (0.65e2.76) ()
0.455 0.941 0.611 0.432
e
e
15.9 35.0 32.4 12.0 4.7
6.12 5.21 2.76 1.62 1.00
0.001 0.002 0.052 0.396
3.47 3.35 1.90 1.48 1.00
Beliefs (personal control, if sick) Disagree (strongly/mod/slightly) Agree (strongly/mod/slightly) Don't know/did not use
18.2 79.1 2.7
0.75 (0.31e1.81) 0.99 (0.43e2.32) 1.00 ()
0.522 0.995
e
e
17.4 80.6 2.0
1.29 (0.29e5.57) 1.67 (0.41e6.82) 1.0 ()
0.733 0.472
e
e
Cognitive style Total score (mean)
e
0.97 (0.94e1.01)
0.117
e
e
e
1.05 (0.97e1.15)
0.197
e
e
Knowledge (spiritual experience) Disagree (strongly/mod/little) Neither agree nor disagree/DK Agree (little/mod/strongly)
57.2 13.0 29.9
1.00 () 2.08 (1.22e3.58) 2.74 (1.52e4.94)
1.00 () 2.17 (1.36e3.44) 2.83 (1.37e5.82)
1.30 (0.81e2.08) 1.14 (0.67e1.99) 1.0 ()
0.275 0.626
e
0.001 0.005
55.6 16.9 27.5
e
0.008 0.001
Openness (to new experiences) Disagree (strongly/mod/little)
6.9
1.00 ()
7.6
1.00 ()
Gender Male Female
% 50.3 49.7
0.44 (0.33e0.58) 1.00 ()
P
