There is a growing global epidemic of obesity and overweight(1). Like in many parts of the world, the escalating obesity problem in Bahrain has reached a level of public heath significance and requires urgent action. Over the last few decades Bahrain has undergone a rapid epidemiological transition, with a reliance on more sedentary jobs and an increase in the availability of high-energy food. This has led to the dramatic rise in the prevalence of obesity. In 2002, the overall prevalence of overweight and obesity was 60 % among adults and 24 % among schoolchildren(Reference Moosa, Gharib, Al-Sairafi, Al-Raees and Al-Amer2, Reference Gharib3). Twenty-three per cent of Bahraini males and 34 % of Bahraini females were obese. This is associated with increased prevalence of non-communicable diseases. CVD and cancer are the leading causes of death(4) and nearly 30 % of Bahrainis aged 40–69 years have type 2 diabetes(Reference Al-Mahroos5).
Obesity is a sequela to an abnormal obesogenic environment and is best tackled with extensive environmental interventions, supported by appropriate policy and legislation(Reference Egger and Swinburn6). However, obesity management is also one of the most cost-effective interventions when combined with regulatory and economic approaches. Clinical management guidelines have been developed with demonstrable efficacy, particularly for dietary counselling, physical activity, behaviour modification and pharmacotherapy(Reference Mercer, Green, Rosenthal, Husten, Khan and Dietz7, Reference Swinburn, Caterson, Seidell and James8). Obese patients counselled to lose weight are more likely to be at a higher stage of readiness for weight loss(Reference Galuska, Will, Serdula and Ford9, Reference Huang, Yu, Marin, Brock, Carden and Davis10).
The primary health care (PHC) system provides the best framework for delivering weight management and physicians have unique professional characteristics which put them in a better position to contribute to the prevention and treatment of overweight and obesity(Reference Campbell, Engel, Timperio, Cooper and Crawford11). The opportunity to deliver lifestyle advice comes from the large percentage of the population who contact their physician(4) and physicians’ extensive involvement with personal prevention(Reference Truswell12). The family practice approach allows physicians to build a rapport, deliver advice tailored to the individual and ensure continuity of care, which provides an ideal set-up for behavioural counselling(Reference Truswell, Hiddink and Blom13). Physicians are the most frequently used source of health information(Reference Truswell, Hiddink and Blom13, Reference Vaandrager and Koelen14) and have been found to be more cost-effective than dietitians in nutritional counselling for obese and other patients(Reference Olsen, Willaing, Ladelund, Jorgensen, Gundgaard and Sorensen15, Reference Pritchard, Hyndman and Taba16).
Despite these facts, PHC is still underutilized for obesity counselling(Reference Mercer, Green, Rosenthal, Husten, Khan and Dietz7) and its capacity is usually restrained by a number of limitations(Reference Glanz17) such as short consultation time, patients’ low motivation and non-compliance, inadequate teaching materials, lack of reimbursement, low level of physicians’ confidence and a shortage of dietitians(Reference Glanz17, Reference Kushner18). Inadequate nutrition knowledge has also been reported by physicians themselves or documented objectively(Reference Murray, Narayan, Mitchell and Witte19, Reference Al-Numair20).
In addition, it has been suggested that obesity management in PHC is deficient mainly due to low identification of patients’ weight status and physicians’ inefficient efforts at intervention(Reference Bramlage, Wittchen and Pittrow21). Less than half of obese individuals are advised to lose weight by their physician(Reference Galuska, Will, Serdula and Ford9, Reference Heath, Grant, Marcheni and Kamps22). Overweight and mild obesity is under-recognized and under-treated unless associated with other medical complications(Reference Heath, Grant, Marcheni and Kamps22, Reference Hebl and Xu23), particularly among the paediatric population(Reference O’Brien, Holubkov and Reis24). Furthermore, health professionals, including physicians, may hold negative attitudes towards obese patients(Reference Puhl and Brownell25, Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26), which may be one of the reasons for the low success rate of obesity management(Reference Ogden, Bandara, Cohen, Farmer, Hardie, Minas, Moore, Qureshi, Walter and Whitehead27).
The present study was conducted to describe the opinions of physicians in Bahrain regarding their role in obesity control, including their capabilities and limitations, and to evaluate their knowledge, attitudes and practices towards obesity prevention and management in PHC. It is concurrent with a proposed obesity-management programme in PHC to empower Bahraini physicians with the necessary professional and structural tools to meet the challenge of the growing obesity epidemic.
Methods
A cross-sectional survey of physicians in Bahrain was conducted in April 2006. The sample size was estimated using the Epi-Info software package version 6·04c (Centers for Disease Control and Prevention, Atlanta, GA, USA) for a 5 % significance level, 10 % precision and a minimum non-response rate of 30 %. For logistical reasons, i.e. the lack of adequate time and resources, a single-stage cluster sampling was conducted. Ten health centres were randomly selected out of twenty-two and all physicians within the selected health centres were included. Another two randomly chosen health centres were added during the period of data collection to account for possible non-responses, increasing the total number of physicians included to 107.
A single self-completed questionnaire was developed based on reviewed literature(Reference Campbell, Engel, Timperio, Cooper and Crawford11, Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Campbell and Crawford28, Reference Fogelman, Vinker, Lachter, Biderman, Itzhak and Kitai29) and was modified further after piloting. Four- and 5-point Likert-scales were used, in which the participants had to indicate their level of agreement with each statement by selecting one from ‘strongly agree’, ‘agree’, ‘neutral’, ‘disagree’ or ‘strongly disagree’; or, in practice statements, from ‘all of the time’, ‘most of the time’, ‘occasionally’, ‘not at all’ or ‘don’t know’.
The questionnaire consisted of five sections; in addition to gathering demographic data, it included eleven statements exploring attitudes and knowledge, three for weight problem identification, nine statements for weight-loss strategies, three statements on self-efficacy and eleven statements on major limitations.
Based on the very low response rate reported in similar studies using postal questionnaires(Reference Campbell, Engel, Timperio, Cooper and Crawford11), it was decided to maximize the response rate by keeping the questionnaire as simple and short as possible and distributing it personally. After ethical approval, the data were collected, coded and entered using the EpiData software version 3·1 (EpiData Association, Odense, Denmark). Analysis was done using the STATA statistical software package version 9 (StataCorp., College Station, TX, USA) with adjustment for the design and the finite population effects to produce the correct standard errors.
Descriptive analysis of the physicians’ attitudes and practices was the primary objective of the study and results are presented mainly as frequencies, adjusted percentages and their 95 % confidence intervals. Means and standard deviations were used to describe the physicians’ personal data.
Because of the small sample size and for ease of presenting the data, most of the responses were grouped together: ‘agreed’ referring to both ‘agreed’ and ‘strongly agreed’, ‘disagreed’ referring to both ‘disagreed’ and ‘strongly disagreed’, and ‘used most of the time’ referring to ‘all’ and ‘most of the time’, v. other responses.
Some of the key outcomes are examined in relation to physicians’ demographic and professional characteristics such as gender, BMI, duration of practice, family medicine certification and attending relevant training courses using univariate and multivariable analyses. However, this should be cautiously interpreted, as identification of predictors was not the primary outcome of the study.
Results
Demographic and professional characteristics
The questionnaire was completed by ninety-seven participants, with a 90 % response rate. The non-respondents were similar to the participants with respect to their gender and qualification. Nearly one-third of physicians were males (n 29) and two-thirds were females (n 68). The age of the physicians ranged from 29 to 59 years. The mean duration of practice in primary care was 11 years. Seventy-eight per cent of the physicians had a degree in family medicine and only 26 % had taken relevant nutrition courses.
Nearly half of the female physicians (49·2 %) and only 13·0 % of the male physicians had a BMI in the normal range. A large proportion of physicians, particularly male doctors, were overweight or obese (Table 1).
Table 1 BMI of the surveyed physicians by gender, Bahrain, 2006
Although male physicians had more years of experience, a larger proportion of female physicians were family medicine graduates; but both had similar training opportunities with regard to nutrition. A significant proportion of junior physicians (those with working experience <10 years) were family medicine graduates and a significantly higher proportion of them had received more training compared with senior physicians (Table 2).
Table 2 Characteristics of the surveyed physicians by gender and length of experience, Bahrain, 2006
*P value for the difference between the two categories of physician, i.e. male and female or junior and senior.
Physicians’ attitudes and knowledge towards obesity management
Ninety-two per cent of the physicians agreed with the statement, ‘Obesity is a big health problem among children and adults in Bahrain which requires urgent action’. Weight-loss counselling was felt to be professionally rewarding among 60·4 % of physicians, who also agreed that their role in obesity prevention and management is unlimited (Table 3).
Table 3 Physicians’ attitudes and knowledge towards obesity management, Bahrain, 2006
Physicians who had nutrition training courses were twice as likely to believe that they can assume a major role in weight management than those who had not, but this was only marginally significant (ORFootnote * = 2·20, 95 % CI 0·98, 4·81, P = 0·055). Nearly two-thirds of the physicians (65·6 %) believed that only a few overweight and obese individuals can successfully lose weight and will be able to maintain the weight loss. Only 36·0 % of physicians reported that the weight management in their practice was effective, irrespective of their gender, length of experience, qualification and training. Nearly 85 % of physicians stated that they would provide advice about weight loss even if the patient had not requested it.
The majority of physicians (88·9 %) would encourage adults with a healthy body weight (BMI = 18·5–24·9 kg/m2) to maintain their weight. More than three-quarters of physicians would not restrict weight management to those who are obese (76·6 %) and four out of five physicians would treat overweight individuals even if no other co-morbidities exist (83·5 %).
Most physicians (96·7 %) believed that even small but sustainable weight losses can produce significant health benefits, with 82·2 % agreeing that this as an important goal to set. This view was particularly held by female physicians (OR = 5·70, 95 % CI 1·87, 17·97, P = 0·003), and a large percentage of physicians (86·8 %) realized that helping patients to adopt a healthy lifestyle is another important goal by itself, irrespective of weight loss.
Obesity recognition
It was found that half of the physicians (50·0 %) occasionally screened their patients for weight problems and only one-third of physicians did this on a regular basis (33·3 %; Table 4). The longer the physician had been in practice, the more likely s/he was to screen for weight problems (OR = 1·70, 95 % CI 1·02, 3·86). Among the 95·6 % of physicians who weighed their patients as part of chronic disease care, 71·4 % did it on a frequent basis.
Table 4 Identification of weight problems by physicians in primary health care, Bahrain, 2006
Weight-management strategies
Nearly all participants included lifestyle advice in their weight-management programme. The majority stated that they advise patients on diet (97·8 %) and physical activity (95·6 %) most of the time (Table 5). Behavioural counselling was used less often, about two-thirds reported using it regularly (64·4 %). Fifty-eight per cent of physicians used leaflets and educational materials most of the time and about a third (32·2 %) used them only occasionally.
Table 5 Weight-management strategies reported by physicians, Bahrain, 2006
Forty-one per cent of physicians referred their patients to dietitians regularly. Involving the patient’s family was reported by more than third of the physicians (39·6 %). Pharmacotherapy and referral to weight-loss surgery were the least used strategies. Only 4·5 % of the practitioners would prescribe medication to treat obese patients and 3·3 % would refer them to surgery if clinically indicated.
Self-efficacy
Physicians felt more confident treating overweight adults than obese adults or overweight/obese children (Table 6). Four out of five physicians (82·4 %) felt professionally prepared to treat overweight patients. In comparison, about two-thirds (68·1 %) of physicians rated themselves as professionally competent to treat obese patients and only half were confident to treat children (50·5 %).
Table 6 Physicians’ reported self-efficacy, Bahrain, 2006
Training courses in obesity management appeared to significantly improve physicians’ confidence in treating obese patients (OR = 4·60, 95 % CI 1·84, 11·44, P = 0·002). Male doctors and trained physicians felt more confident in treating children (OR = 3·20, 95 % CI 1·39, 7·42, P = 0·008 and OR = 1·80, 95 % CI 0·95, 3·39, P = 0·069, respectively).
Barriers to weight management
The three major obstacles identified by physicians as the most limiting in delivering optimal care to obese patients were short consultation times (91·0 %), a lack of weight-management clinics (81·1 %) and the absence of national weight-management guidelines (78·6 %; Table 7).
Table 7 Physicians’ reported major limitations in weight-management practice, Bahrain, 2006
Some physicians (71·1 %) agreed that an inadequate number of dietitians is a limiting factor. A similar proportion (70·0 %) found that low patient motivation and poor compliance to weight-loss interventions are important barriers in weight management.
Inadequate training in providing dietary and lifestyle counselling for obese patients was stated by nearly two-thirds of physicians as a reason why they felt less efficient in providing optimal care (64·4 %). However, only 30·0 % of the physicians felt that inadequate knowledge was a potential barrier to providing obesity management. This response did not differ by gender, BMI, duration of practice, qualification or training.
The lack of availability of weight-loss drugs was reported by more than half of the physicians (57·8 %) as a limiting factor for successful weight management. The majority of physicians (91·2 %) did not consider the low success rate in obesity management as a limitation. This view was supported particularly by physicians with fewer years of practice (OR = 12·50, 95 % CI 1·60, 96·64, P = 0·017).
A large number of physicians (78·0 %) did not consider a lack of interest in obesity care as being an obstacle, particularly male (OR = 2·80, 95 % CI 1·06, 7·47, P = 0·039) and normal-weight physicians (OR = 2·80, 95 % CI 1·37, 5·71, P = 0·006).
Discussion
From the present survey it is obvious that physicians were aware of the magnitude of obesity as a major public health problem in Bahrain. In fact, a large proportion of the physicians held strong positive views regarding their role in obesity control as reflected by their strong obligation and commitment to advice on weight loss even if not asked for by their patients. This is despite the feeling of a low level of effectiveness in their weight-loss practices and their negative views regarding the success of obese and overweight patients to lose and maintain the lost weight. The reasons for the existing disparity between the level of interest shown by physicians towards obesity management and their perceived low level of effectiveness is not clear; whether it results from real experience or from published literature on long-term obesity treatment(Reference Foster, Wadden, Makris, Davidson, Sanderson, Allison and Kessler30, Reference Levine, Wigren, Chapman, Bergman and Rivlin31) remains to be established.
The physicians surveyed demonstrated sufficient knowledge regarding the indication to deliver weight-control counselling and the definition of weight-loss goals(Reference Goldstein32). The majority of physicians were aware of the importance of setting realistic weight-loss goals and the adoption of a healthier lifestyle. They also acknowledged the importance of early intervention even in the absence of co-morbidities. However, a noticeable gap exists between the physicians’ attitudes and knowledge regarding weight management and their current practices. This is demonstrated by the fact that only one-third of respondents would identify their patients’ weight problems on regular basis despite that the majority agreed to advice about weight loss spontaneously, similar to what has been found in an earlier study(Reference Brotons, Ciurana, Piñeiro, Kloppe, Godycki-Cwirko and Sammut33). Physicians also seemed to identify weight problems as part of chronic disease care more often than for screening purposes. This suggests a therapeutic rather than preventive approach towards weight problems.
Physicians utilized various approaches in weight-loss management. Similar to other studies(Reference Campbell, Engel, Timperio, Cooper and Crawford11, Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Fogelman, Vinker, Lachter, Biderman, Itzhak and Kitai29, Reference Cade and O’Connell34), lifestyle therapy was the most frequently used approach. The benefits of long-term behavioural modification and family involvement in the management of obese patients have been clearly highlighted in the literature(Reference Rippe, McInnis and Melanson35, 36). A large proportion of Bahraini physicians valued strategies that support lifestyle changes and reported using behavioural counselling or involving family members on a regular or occasional basis. This far exceeds what has been reported by other physicians(Reference Campbell, Engel, Timperio, Cooper and Crawford11, Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Fogelman, Vinker, Lachter, Biderman, Itzhak and Kitai29).
Pharmacotherapy was used less often even if clinically indicated. This is probably because drugs are expensive; or because of concerns about its safety or possible ineffectiveness. The latter is unlikely, as the majority perceived the unavailability of anti-obesity drugs in governmental clinics as a potential limitation in obesity management.
Despite its long-term cost-effectiveness, surgery is less utilized, even in developed countries. This may be due to physicians being less aware of its potential benefits, the inaccessibility to qualified surgeons or fear of possible risks(Reference Foster, Wadden, Makris, Davidson, Sanderson, Allison and Kessler30).
The major constraints identified by physicians to provide obesity management were mainly related to the PHC system and/or factors related to the patient. Short consultation time is a well-recognized barrier to preventive services in PHC and has been reported by the majority of physicians in other studies(Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Roland, Bartholomew, Courtenay, Morris and Morrell37). The absence of clinical guidelines was reported as an important limiting factor; however, in some developed countries where national guidelines exist, only a few practitioners demonstrated good awareness of them(Reference Truswell12, Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26). Moreover, low patient motivation and patient non-compliance were considered as limiting factors for the success of weight-loss programmes by a large proportion of physicians, as reported by others(Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26).
Similar to earlier studies(Reference Campbell, Engel, Timperio, Cooper and Crawford11, Reference Foster, Wadden, Makris, Davidson, Sanderson, Allison and Kessler30), Bahraini physicians felt more confident to treat overweight than obese adults but felt less competent to treat overweight children than other surveyed physicians(Reference Jelalian, Boergers, Alday and Frank38).
The present study has several limitations. The use of cluster sampling might underestimate the actual variance; this is because participants within the same health centre are more likely to share similar experiences. However, this design effect was taken into account during the analysis.
The questionnaire was kept short to increase the response rate, but this limited the scope of the information that could be gathered. For example, the knowledge section focused on two general aspects only: the indication for intervention and weight-loss goals. Other surveys have overcome this by using separate knowledge and practice questionnaires(Reference Campbell, Engel, Timperio, Cooper and Crawford11).
Furthermore, weight measurement was used as an indicator of identification of weight problems. This may have overestimated the actual practice as it is difficult to assume with certainty that measured weight was used to calculate BMI and was followed by the appropriate action. This should be considered when comparing these findings with results from developed countries, whose physicians reported measuring BMI less frequently(Reference Brotons, Ciurana, Piñeiro, Kloppe, Godycki-Cwirko and Sammut33).
The study was based entirely on self-reporting. Consequently, it is possible that certain practices may have been over-reported. This can be validated by direct observation or auditing patients’ records. Moreover, the study investigated the frequency that each approach was utilized, but this does not take into account the quality of advice given.
However, the present study is the first one in the region to look at physicians’ attitudes, knowledge and practices towards the issue of obesity management and has several implications for clinicians and policy makers. It runs contemporaneously to the launching of a structured obesity-management practice in Bahrain. It will help highlight opportunities for interventions and identify barriers that need to be tackled in order to deliver high-quality obesity management in the PHC system. The study’s main strengths come from the high response rate and the relatively large sample size, as it captured 56 % of the total population of physicians, thus minimizing selection bias and increasing precision.
To improve obesity management within the PHC system in Bahrain, the organizational structure needs to be considered seriously. The following may be suggested from the findings of the current survey.
1. The consultation time allocated for counselling and treating overweight patients needs to be increased, whether this is integrated within the same practice or delivered through special clinics.
2. Physicians should be supported with clear guidance on obesity management using evidence-based clinical guidelines, with clear referral protocols to surgery or other health professionals.
3. Physicians have a great motivational role but are more restrained by their busy clinics, which makes collaboration with other health professionals a priority that should be considered without hesitation. There should be a focus on recruiting more dietitians and/or other trained health workers who can deliver in-depth counselling and provide regular long-term follow-up, either attached to physicians’ practices or working independently through easily accessible specialty clinics.
4. Nutrition education needs to be integrated within medical school curricula and postgraduates’ continuing medical education.
5. The role of surgery and pharmacotherapy in obesity management needs to be emphasized. Identifying well-trained surgical teams to which to refer patients and providing licensed anti-obesity drugs at lower price are important steps to be considered to ensure comprehensiveness of care.
6. Training on lifestyle counselling and behavioural modification skills is a necessity acknowledged by physicians in the present study and has been found to improve the frequency and quality of counselling delivered to obese patients(Reference Smikin-Silverman and Wing39), decrease negative attitudes towards tackling the problem of obesity and enhance physicians’ confidence(Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Wiese, Wilson, Jones and Neises40, Reference Katz, Feigenbaum, Pasternak and Vinker41), particularly in counselling parents of overweight children.
7. Our physicians need more emphasis on their preventive role in obesity. They need to translate their knowledge about the importance of early identification of weight problems into actual practice.
8. Other opportunities for more effective weight management in PHC exist and need to be discussed; these include patient registers, designated physician specialists, supervised group therapy(Reference Renjilian, Perri, Nezu, McKelvey, Shermer and Anton42) and subsidized referral to commercial slimming groups(Reference Maryon-Davis43).
Our study has identified areas that need to be explored further using qualitative studies. The physicians’ feeling of ineffectiveness in treating obese individuals despite their high level of interest, good amount of knowledge and the reported high utilization rate of various weight-loss approaches needs to be explored. This could be explained partly by the fact that physicians have minimal influence on patients’ environment, a major determinant of the obesity epidemic. In addition, the physicians’ strong perception of the external obstacles and under-reporting of their professional limitations need to be ascertained and explained further.
Because of the high prevalence of overweight among physicians, their opinions regarding the concept of being a role model, as suggested by physicians in other studies(Reference Bocquier, Verger, Basdevant, Andreotti, Baretge, Villani and Paparonis26, Reference Cade and O’Connell34), and the positive effect of personal experience on weight management also need to be discussed with the physicians.
Conclusion
The present study is the first one on physicians’ knowledge, attitudes and practices towards obesity prevention and management in the region, and is concurrent with the launching of the obesity clinic project in PHC in Bahrain.
It is encouraging that physicians in Bahrain are aware of the escalating obesity epidemic and are willing to participate responsibly. They had a greater sense of professional satisfaction than practitioners in other countries(Reference Campbell, Engel, Timperio, Cooper and Crawford11), probably because they set realistic weight-loss goals. Despite this, they have judged their weight-management practice as ineffective. This could be due to their negative attitude towards obese patients and the success of weight management, as the majority believed that obese patients are unlikely to achieve and maintain weight loss or to comply with weight-loss strategies. In addition, they have identified a number of organization-related barriers as the major limitations in their practice, but they were less critical in appraising their knowledge and skills or addressing their negative view towards the success of weight-loss programmes.
Our physicians are well informed about the health benefits of modest weight loss. Although they have reported measuring their patients’ weight more often than other physicians, it is not known to what extent this reflects actual identification of weight problems.
Physicians have utilized various approaches in weight-loss management. They have reported involving patients’ families and utilized behavioural modification to a greater extent than physicians in other developed countries. However, in the absence of objective assessment, the frequency of their practices could have been over-reported.
The importance of collaboration with other health professionals, especially dietitians, is well recognized, but heavily under-resourced. Anti-obesity drugs and surgery are underutilized even if indicated, which should be considered in future nutrition education. Like other studies, training in lifestyle counselling and behaviour modification was highlighted as an important need, and seems to predict physicians’ confidence in implementing weight-management strategies.
Acknowledgements
Sources of funding:None. Conflict of interest declaration:There were no conflicts of interest. Author contributions:A.A.-G. collected and analysed the data, and wrote the report; R.U. supervised and reviewed the survey. Acknowledgements:The authors extend special thanks to the primary care physicians who participated in the survey, to Dr Phil Edward for the great help offered in sampling and analyses, and to Mrs Claire Schofield for her continuous support.
