• Attitudes;
  • health professionals;
  • overweight;
  • weight management


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

This systematic review aims to address the question of whether health professionals' weight status is associated with attitudes towards weight management. Twelve eligible studies were identified from a search of the Cochrane Library, Ovid MEDLINE, EMBASE, PsycINFO, CINAHL and three Chinese databases, which included 14 independent samples comprising a total of 10 043 respondents. Attitudes towards weight management were classified under eight attitude indicators. Quantitative synthesis of the findings of included studies showed that health professionals of normal weight were more likely to be more confident in their weight management practice, perceive fewer barriers to weight management and have more positive outcome expectations, have stronger role identity and more negative attitudes towards obese individuals than health professionals who were overweight or obese. However, there was no difference between overweight and non-overweight health professionals in their perceptions of the causes and outcomes of obesity. In addition, being female and having relevant knowledge and clinical experience of weight management appeared to predict positive attitudes towards obesity/obese patients and high self-efficacy in weight management, respectively. Future research should focus on prospectively theory-driven studies, and employ appropriately validated instruments and multivariate analyses to identify the relative contribution of weight status to attitudes towards weight management.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

The evidence on weight-based disparities in healthcare settings through recently updated reviews (1–4), have demonstrated that health professionals (e.g. physicians, nurses, psychologist and dietitians) share the prevailing social biases that the public hold against obese individuals, including beliefs that obese patients are lazy, non-compliant, undisciplined and have low will-power.

Meanwhile, health professionals, like much of the general population, are not immune to overweight or obesity, although the prevalence of this would appear to be lower than the public (5–8). The questions addressed in this review are whether health professionals who themselves struggle with excess weight, differ from their non-overweight colleagues in their views of overweight/obesity or overweight/obese individuals, perceptions of themselves as role models for their patients and their confidence in influencing overweight or obese patients.

That health professionals' own health and health habits may influence their attitudes towards relevant professional behaviours has been demonstrated by many studies of cigarette smoking and physical exercise. Health professionals who smoke have less favourable attitudes towards smoking cessation (9–18), less confidence in their efficacy to help the patients to quit smoking (6,19–26) and less intention to counsel their patients (19,27,28), and are less likely to perceive themselves as role models for their patients or the general population (13,29–36), when compared with non-smoking health professionals. Similar findings have been reported with respect to physical activity with unfit health professionals being less likely to hold positive attitudes towards promoting physical activity than physically fit health professionals (37–41). Given these associations between health professionals' personal health behaviours and their attitudes towards health promotion it seems possible that overweight or obese health professionals will hold less positive attitudes towards weight management than their non-overweight counterparts. Thus the focus of this systematic review is to examine the relationship between health professionals' weight status and attitudes towards weight management.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

Eligibility criteria

Studies selected for inclusion in this review met the following criteria. They had to:

  • 1
    Report and categorize weight status (normal weight, overweight or obesity) of health professionals (doctors, nurses, dietitians, physiotherapists or psychologists);
  • 2
    Report individuals' attitudes towards aspects of weight management, i.e. towards (i) overweight/obesity; (ii) overweight/obese individuals and (iii) weight treatment;
  • 3
    Report statistical relationships between (1) and (2);
  • 4
    Be published in English or Chinese.

The inclusion criteria were not restricted by types of studies or participants. Studies of patients or the general population as recipients of weight management, were also included if the association between health professionals' weight status and relevant attitudes towards weight management were assessed.

Studies were excluded if:

  • 1
    The data on health professionals were not reported independently, but were e.g. mixed with data from unqualified staff (e.g. medical students, nursing students and research staff);
  • 2
    The data regarding weight status or related attitudes measured, remained unclear or incomplete after reading the full text of the paper;
  • 3
    Health professionals' weight status was analysed as a continuous variable associated with related attitudes, as the aim of this review was to compare attitudes towards weight management between overweight and non-overweight health professionals.

Search strategies

Electronic databases of Ovid MEDLINE (1948–2010.6), EMBASE (1947–2010.6), PsycINFO (1806–2010.6), CINAHL (1982–2010.6) and Cochrane Library were searched by one researcher (DZ) with guidance on the search strategy from the other two authors. A sensitive search strategy was devised with which involved using Boolean operators to combine the terms (‘doctor’ OR ‘dietitian’ OR ‘nurse’ OR ‘physiotherapist’ OR ‘psychologist’) AND ‘weight’ AND ‘attitude’. Specific search terms used are showed in Table 1. The following Chinese databases were also searched: Chinese Biomedical database (CBM, 1978–2010.6), VIP Chinese Science Journals Database (Vip CSJD, 1989–2010.6) and China Academic Journals Full-text Database (CAJFD, 1994–2010.6). The reference lists of all full-text papers retrieved were also checked manually to ensure that additional papers were not missed. No attempt was made to access unpublished studies or other ‘grey’ literature.

Table 1.  Search terms
FacetsSearch terms
  • *

    Potential members of multidisciplinary obesity treatment programmes.

Health professionals*DoctorPhysician/s; Medical Staff/s
DietitianDietitian/s; Dietician/s; Nutritionist/s
NurseNurse/s; Nursing Staff/s
PhysiotherapistPhysiotherapist/s; Physical therapist/s; exercise physiologist/s
PsychologistPsychologist/s; Psychiatrist/s
WeightBody Weight; Obesity; Overweight; Body Mass Index/BMI; Body Size; Waist Circumference; Waist-Hip Ratio; Body Fat; Body Image; Body Composition
AttitudeAttitude/s; Belief/s; Opinion/s; Perception/s ; View/s

All titles and abstracts retrieved by electronic searching were downloaded to the reference database EndNote; duplicates were removed. Initial screening was undertaken by one researcher (DZ) for all English language papers and then checked by another (IJN). This was not possible for Chinese papers because only one researcher (DZ) could read Chinese. However, the Chinese researcher (DZ) was asked by a second researcher to provide a verbal summary of the paper and a rationale for their screening decision. Where it was not possible to exclude articles based on title and abstract, full-text versions were obtained and their eligibility was assessed. All full-text articles were examined, and disagreements about inclusion were discussed until agreement was reached.

Data analyses

Data about authors, year of publication, country, study design, sampling, sample size and setting, instruments, weight and attitude variables measured and main results were abstracted from the selected studies by one researcher (DZ), and reviewed or verified by the other two authors; these data are summarized in Tables 2 and 3.

Table 2.  Summary of basic characteristics of 12 included studies
AuthorLocationDesignSamplingSample size and settingResponse rate (%)Instruments (questionnaire)Comments
  • *

    A very small number of paediatricians responded by telephone interview.

  • Cross-sectional survey (CS).

  • Random sampling (RS), non-random sampling (Non-RS), random grouping (RG), non-random grouping (Non-RG).

  • §

    Researcher-developed (R), existing (E).

  • **

    Delivery by Post/mail (P), Telephone (T), Website (W), Hand (H).

  • ††

    Self-administered (SA), interviewer-administered (IA).

  • AAP, American Academy of Paediatrics; ADA, American Dietetic Association; CARNA, College and Association of Registered Nurses of Alberta; FHSA, Family Health Services Authorities; GPs, general practitioners; NAPNP, National Association of Paediatric Nurse Practitioners; NCPS, North Carolina Paediatrics Society; PNPs, paediatric nurse practitioners; PNs, primary nurses; RDs, registered dietitians; RNs, registered nurses, RR, response rate.

Watson, et al. (51)CanadaCSRS626 RNs in a random sample from the database of the CARNA46RTested (clearly reported)PSAA provincially representative sample
Theoretically based study
The instrument with acceptable psychometric properties
Bocquier, et al. (47)FranceCSRS600 private GPs in a randomly stratified sample from Provence in southern France56RNo formal psychometric testingTIAA regionally representative sample
Brotons, et al. (6)SpainCSRS2082 GPs in a random sample from 11 European national colleges of GPs54 (50–65)RNo formal psychometric testingPSAA multinational sample
Hoppe, et al. (52)UKCSRS586 PNs in a random sample from 10 randomly selected FHSAs65RTested (inadequately reported)PSAA moderate sample
Theoretically based study
Foster, et al. (53)USACSRS620 family physicians in a random samples of members of the New Jersey Academy of Family Physicians13RNo formal psychometric testingPSAA moderate sample
Low RR
Hendershot, et al. (54)USACSNon-RS2629 elementary school nurses from the National Association of School Nurses42RNo formal psychometric testingWSAA large sample across the country
Theoretically based study
Miller, et al. (7)USACSRS760 RNs in a random sample from six representative states across the country16Not reportedNot reportedPSAA moderate sample
Low RR
Perrin, et al. (50)USACSNon-RS355 paediatricians from both members of the NCPS and the AAP in North Carolina62 (unadjusted)RNo formal psychometric testingPSAA small sample
71 (adjusted)Relatively comprehensive analysis
Story, et al. (55)USACSRS202 paediatricians in a random sample from the AAP19RNot reportedP*SAInclusion of three professions from across the country
444 RDs in a random sample from two practice groups of the ADA27
293 PNPs in a random sample from the NAPNP33Low RR
Warner, et al. (56)USACSNon-RS477 military family physicians from the Uniformed Service Branch of the Academy of Family Physicians40ENot reportedWSAA small sample Low RR
Hash, et al. (48)USAA quasi-experimental post-test designNon-RG226 patients in five physician offices RTested (inadequately reported)HSAA small sample
Group one: 112 patients visiting two obese physiciansSome confounders from patients and physicians could introduce bias
Group two: 114 patients visiting three non-obese physicians
Hick, et al. (49)USAA quasi-experimental post-test designRG143 participants in an university randomly assigned into: ENo formal psychometric testingHSAA small sample A small number of demographic variables compared between two groups
Group one: 72 participants viewing a weight-appropriate nurse image
Group two: 71 participants viewing an overweight nurse image
Table 3.  Summary of weight variables and related attitude variables analysed by 12 included studies
AuthorCategories of weight variablesVariables of attitudes towards weight management and attitude indicatorsUnivariate analysisMultivariate analysis
  • Studies appeared more than once if analysis varied by subsamples, weight variables or attitude indicators.

  • Blank cells mean no examination of indicators.

  • *

    Weight variables were significantly associated with one item of two items.

  • Significant difference was found between normal weight and obese nurses.

  • BCO, beliefs about the causes of obesity; BMI, body mass index; BPCOP, beliefs about the personality characteristics of obese persons; ER, emotional response; GPs, general practitioners; NA, not available; NS, no significance; OE, outcome expectations; OR, odds ratio; PB, perceived barriers; PNPs, paediatric nurse practitioners; PNs, primary nurses; PRI, professional role identity; PS, perceived severity; RDs, registered dietitians; RNs, registered nurses; SE, self-efficacy; VAS, visual analogue scale.

Watson, et al. (51)RNs' self-reported weight and height: BMI (<25, ≥25)ER  
 Negative reflection to obese patients ( a total score of 14 items, VAS)= 0.001
 Stereotypic characteristics of obese patients (a total score of 2 items, VAS)< 0.001
 Characteristics of obese individuals (a total score of 9 items, VAS)NS
 Controllable factors contributing to obesity (a total score of 8 items, VAS)< 0.001
 Supportive roles in caring for obese patients (a total score of 3 items, VAS)NS
Bocquier, et al. (47)GPs' self-reported weight and height: BMI (<25, 25–29.9, ≥30)SE = 0.05, (<25 vs. 25–29.9): OR = 1.69 (1.11, 2.56); (<25 vs. ≥30): OR = 1.41 (0.47, 4.17)
 Feeling effective in management of weight problemsSig. (P value: NA)
 Thinking obese people tend to be lazier and more self-indulgent than normal weight peopleNS
Brotons, et al. (6)GPs' self-reported weight and height: BMI (≤30, >30)SENS 
 Perception of effectiveness in helping patients to achieve or maintain normal weight (yes vs. no)
Hoppe, et al. (52)PNs' self-reported weight and height: BMI (< 23 [median], ≥ 23 [median])OE  
 Preventability of obesity (1–7)< 0.01
 Treatability of obesity (1–7)NS
 Benefits of weight loss (1–7)NS
 Failed weight lose due to patients' non-compliance (1–7)NS
 Failed weight lose due to patients' poor motivation (1–7)NS
 Failed weight lose due to inadequate weight loss methods (1–7)NS
 Biological causes (1–7)NS
 Lifestyle causes (1–7)NS
 Seriousness of obesity (1–7)NS
 Cardiovascular consequences of obesity (1–7)NS
 Non-cardiovascular consequences of obesity (1–7)NS
 Perceived confidence in weight loss counselling (1–7)NS
 Perceived confidence in patients' following the advice (1–7)NS
 Perceived confidence in patients' losing weight (1–7)NS
 Perceived success in bringing about weight loss (1–7)NS
Foster, et al. (53)Physicians' self-reported weight and height: BMI (<25, ≥25)ER  
 Emotional reactions to the appearance of obese patients (4 items, 1–5)NS
 More time working on weight management if reimbursed appropriately (1 item, 1–5)= 0.002
 Educating patients on the health risks of obesity (1 item, 1–5)= 0.006
 Being role models by maintaining a normal weight (1 item, 1–5)NS
 Medications to treat obesity (2 items, 1–5)= 0.0007*
 Achieving a normal weight and improving obesity-related health complications (6 items, 1–5)NS
 Personal characteristics of obese individuals (9 items, 1–7)NS
 Genetic, environmental, social and psychological cause of obesity (11 items, 1–5)NS
 Consequences of obesity (2 items, 1–5)NS
 Perceived competence and success (3 items, 1–5)NS
Hendershot, et al. (54)Nurses' self-reported weight and height:SE  
 Underweight (BMI ≤ 18.5) Confident in ability regarding measurement of BMI (a score of 10 items, 1–5)NS, = 0.254
 Normal weight (BMI: 18.5–24.9)OE 
 Overweight (BMI: 25–29.9) Six activities will help maintain a healthy weight BMI (a score of 6 items, 1–5)NS, = 0.917
 Total number of perceived benefits of measuring BMINS, = 0.417
 Obese (BMI ≥ 30)PB 
 Total number of PB for measuring BMI= 0.002
Miller, et al. (7)RNs' self-reported weight and height: BMI (<25, ≥25)SE  
 Feeling competent to provide obesity education to patients (no reporting of the measurement scale)NS, = 0.315
Perrin, et al. (50)Paediatricians' self-perception weight:SE < 0.05, OR = 5.69 (2.30, 14.10)
 ‘thin’ vs. ‘average’ Perceived difficult in obesity counselling (more difficult vs. easier and no difference)< 0.05, OR = 5.13 (2.25, 11.7)< 0.05, OR = 3.84 (1.11, 13.30)
 ‘overweight’ vs. ‘average’NS, OR = 2.18 (0.94, 5.10)
Paediatricians' self-reported weight and height: BMI (<25, ≥25)SE  
 Perceived difficult in obesity counselling (more difficult vs. easier and no difference)NS, OR = 1.49 (0.75, 3.03)NS, OR = 1.61 (0.48, 5.56)
Story, et al. (55)Paediatricians' self-reported weight and height: BMI (<25, 25–29.9, ≥30) RDs' self-reported weight and height: BMI (<25, 25–29.9, ≥30) PNPs' self-reported weight and height: BMI (<25, 25–29.9, ≥30)OE  
 A need for treatment of Childhood/adolescent overweight (2 items, 1–5)NS
 Overweight children/adolescents will outgrow their overweight (2 items, 1–5)NS
 Childhood/adolescent overweight is more amenable to treatment (2 items, 1–5)NS
 The negative impact of child and adolescent obesity on health outcome (2 items, 1–5)NS
 Barriers to effective treatment (9 items, 1–5)NS
 Perceived skill level in obesity management (7 items, 1–3)NS
Warner, et al. (56)Physicians' self-reported weight and height: BMI (<25, ≥25)SE  
 Perceived gratification from counselling ( 1–5)NS
 Recommend weight loss when present with another problem (1–5)NS
 Feeling comfortable about counselling patients (1–5)NS
 Patients can maintain an ideal weight ( 1–5)NS
 Patients would not lose weight based on only decreased caloric intake (1–5)NS
 Stereotypical characteristics of obese patients (3 items, 1–5)NS
 The obligation to counsel obese patients (1–5)NS
 A role model for patients by maintaining a normal weight (1–5)= 0.019, OR = 12 (1.51, 95.63)
Hash, et al. (48)Physicians' self-reported weight and height and BMI: BMI (<30, ≥30)SE  
 Patients' perceived confidence for receiving health advice from physicians (scale scores)= 0.038
 Patients' perceived confidence for receiving advice for treatment of illness (subscale scores)= 0.049
 Patients' perceived confidence for receiving advice for weight and fitness (subscale scores)NS, = 0.075
Hick, et al. (49)Nurses (images): Body size (normal weight, overweight)SE  
 Public confidence in nurses' ability to provide diet and exercise counselling (VAS: 0–10)= 0.006

Examination of the included studies showed that different terms were used to define cognitive variables (i.e. beliefs, attitudes, perceptions and views). To facilitate interpretation we refer to all cognitive variables examined in this review as ‘attitudes’ on the grounds that the term of ‘attitude’ was used most frequently by the authors of the included articles and attitude can be considered the sum of beliefs and usually measured by assessing a person's belief (42).

Of the 12 included studies, 10 categorized weight variables according to the World Health Organization's classifications of body mass index (BMI) for normal weight, overweight or obesity. However, meta-analysis was not attempted because the studies differed markedly in the attitudes measured and the types of health professional sampled. Instead, we undertook a semi-quantitative approach to data synthesis, similar to that adopted by recent reviews of obesity (43,44), to evaluate associations between health professionals' weight status and attitudes towards weight management.

To classify all studied attitudes and enable comparison and synthesis of the data across studies included, eight attitude indicators were extracted from 12 theoretical domains defined by Michie et al. (45) and adopted by Godin et al. (46) as an integrative framework in their systematic review. These attitude indicators were:

  • 1
    Beliefs about the causes of obesity (BCO) defined as perceptions of the factors contributing to obesity, including external uncontrollable (genetic factors) and internal controllable factors (environmental factors);
  • 2
    Beliefs about the personality characteristics of obese persons (BPCOP) defined as perceptions of personal characteristics of obese individuals, including positive (e.g. industrious, pleasant and graceful) and negative components (e.g. lazy, unpleasant and awkward);
  • 3
    Emotional response (ER) defined as an intrapsychic reaction towards obese patients, including positive (e.g. feeling empathy for obese patients) and negative (e.g. feeling uncomfortable when examining obese patients) components;
  • 4
    Outcome expectations (OE) defined as personal judgment that a particular behaviour related to weight management will produce a certain consequence, including positive (e.g. anticipated weight loss if restriction of diet) and negative (e.g. weigh loss is impossible for most obese patients) components;
  • 5
    Perceived barriers (PB) defined as beliefs about the material and psychological costs of practising weight management (e.g. lack of time, reimbursement and inadequate resources);
  • 6
    Professional role identity (PRI) defined as an individual's perception of “self” in relation to their professional role in the management of obesity and overweight people (e.g. being role models by maintaining a normal weight);
  • 7
    Perceived severity (PS) defined as feelings concerning the seriousness of obesity or of leaving it untreated (e.g. cardiovascular consequences of obesity);
  • 8
    Self-efficacy (SE) defined as personal perception of one's confidence or capability to carry out weight management (e.g. perceived competent in educating obese patients).

To summarize the data from the studies, we identified four possible types of association between weight status and each attitude indicator: inverse, statistically significant association (P < 0.05) (i.e. those of lower weight have more positive attitudes towards weight management than those of higher weight); positive, statistically significant association (P < 0.05) (i.e. those of lower weight have less positive attitudes towards weight management than those of higher weight); no significant association, characterized by the absence of any statistically significant association between weight status and an attitude indicator; and mixed association characterized by a combination of inverse and positive associations (i.e. those of middle weight have less or more positive attitudes towards weight management than those who are of heavier or lighter weight).


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

Identification of studies

The results of the search and screening of papers are summarized in Fig. 1. After 101 duplicate records were removed, an initial 902 abstracts were identified and screened, of which the majority (n = 873) did not meet the eligibility criteria. As a result, 29 articles were retained for more detailed evaluation based on reading the full text, with a further paper identified from reference lists. Of 30 studies in total, 12 studies met all the eligibility criteria and were included in this review.


Figure 1. Flow diagram of search and results.

Download figure to PowerPoint

Description of included studies

The characteristics of the 12 included studies are summarized in Table 2. These studies originated from the USA (n = 8), Canada (n = 1), France (n = 1), Spain (n = 1) and UK (n = 1); no eligible studies originated from China. Eleven papers were published in 2002 or later; one paper was published in 1997.

Ten studies used cross-sectional surveys to examine the weight–attitude association in 12 independent samples comprising a total of 9674 health professionals: six (sub)samples of 4336 doctors, five (sub)samples of 4894 nurses and a subsample of 444 dietitians. None of the studies sampled psychologists or physiotherapists. Two studies used quasi-experimental post-test designs and independently investigated 226 patients and 143 members of the public.

The instruments employed were all self-reported questionnaires, more than half of which elicited relevant attitudes by multi-item Likert scales. Of 12 studies, only one study clearly and completely reported how the questionnaire was developed and tested; neither of the two studies using existing questionnaires reported previous reliability and validity.

Association between weight variables and attitude indicators

Table 3 summarizes weight variables, attitude variables and related statistical results. The types of weight–attitude association from the 12 included studies are listed in Table 4. All eight attitude indicators were examined either in samples of doctors or nurses and four indicators were examined in dietitians. In total, the 12 studies reported 40 univariate analyses of the weight–attitude association and eight multivariate analyses. Twelve of the univariate analyses reported statistically significant findings and three of the multivariate analyses were also significant.

Table 4.  Summary of the associations stratified by specialty and attitude indicators
Attitude indicatorsUnivariate analysisMultivariate analysis*
Numbers of examinationNumbers of inverse associationNumbers of positive associationNumbers of mixed associationNumbers of no associationNumbers of examinationNumbers of inverse associationNumbers of mixed associationNumbers of no association
  • Blank cells mean no examination of indicators.

  • Studies appeared more than once if analysis varied by subsamples, weight variables or attitude indicators.

  • *

    No positive association was reported.

  • BCO, beliefs about the causes of obesity; BPCOP, beliefs about the personality characteristics of obese persons; ER, emotional response; OE, outcome expectations; PB, perceived barriers; PRI, professional role identity; PS, perceived severity; SE, self-efficacy.

 BPCOP (pos.)200021001
 ER (pos.)100011001
 OE (pos.)210011001
 BPCOP (pos.)20101    
 ER (pos.)10100    
 OE (pos.)42002    
 BPCOP (pos.)         
 ER (pos.)         
 OE (pos.)10001    
 BPCOP (pos.)401031001
 ER (pos.)201011001
 OE (pos.)730041001

Self-efficacy was most frequently examined by 11 studies with 13 univariate analyses and four multivariate analyses. An inverse weight–SE association was identified by three studies (47–49) and a mixed association by one study (50). No studies reported a positive association between weight status and SE.

Of the three studies reporting an inverse association, a survey of a random sample of 600 private general practitioners (GPs) revealed that normal weight (BMI < 25) was significantly associated with a feeling of effectiveness in weight management in the multiple logistic regression (47). This evidence was further reflected in two quasi-experimental studies which reported that non-obese physicians (48) or weight-appropriate nurses (49) may inspire patients to be more confident in them as health educators.

A mixed association was identified in a convenience sample of 355 paediatricians relating to self-perceived weight status but not BMI classifications (50) with self-perceived ‘thin’ (odds ratio = 5.69, 95% confidence interval [CI]: 2.30–14.1) or ‘overweight’ (odds ratio =  3.84, 95% CI: 1.11–13.3) being associated with greater perceived difficulty in counselling about weight than self-perceived ‘average’. This study suggested that self-perceived weight status may be a stronger determinant of perceived difficult in obesity counselling than BMI classifications.

Outcome expectations

Outcome expectations was the second most frequently examined. Three of six studies found an inverse association between positive OE and the weight status (51–53), with no other type of association being found.

A study of a random sample of 586 UK practice nurses (52) found that nurses with low BMIs rated obesity as more preventable (P < 0.01) than those nurses who were overweight. Similarly, another study of a random sample of 626 Canadian nurses (51) indicated that thinner nurses (BMI < 25) were more likely to believe obesity to be controllable (P < 0.001). This significant inverse association between positive OE and weight status was also reflected in a study of a random sample of 620 family physicians (53).

Perceived barriers

Two of three studies with five univariate analyses identified a positive weight–PB association (53,54), but a study of three subsamples (55) did not.

A study of a sample of 2629 school nurses found that personal weight status affected PB in measuring BMI for school children with normal weight nurses (BMI, 18.5–24.9) reporting significantly fewer barriers than obese nurses (BMI ≥ 30) (54). A survey of 620 family physicians also reported similar findings (53).

Beliefs about the personality characteristics of obese persons

One of five studies of the association between positive BPCOP and weight status (47,51–53,56) found a positive association in a random sample of 626 Canadian registered nurses (RNs) (51). This study suggested that nurses in lower weight categories (BMI < 25) showed less positive attitudes towards obese patients (P < 0.001).

Professional role identity

Of three studies examining the weight–PRI association, two studies of physicians (53,56) identified a significant inverse association while a study of RNs did not (51).

A study with a random sample of 620 family physicians (53) found that physicians with BMI < 25 were more likely to feel obligated to educate patients on risks for obesity than those with BMI ≥ 25 (P = 0.006). Also, a convenience sample of 477 military family physicians (56) reported that normal weight physicians were 12 times (95% CI: 1.51–95.63, P = 0.019) more likely to believe that they were a role model for patients than physicians who were overweight or obese.

Emotional response

One of three studies examining the weight–ER association in a sample of RNs (51) found that thinner nurses were less positive in their response to obese patients than overweight nurses. However, this association was not identified by two samples of doctors (53,56).

Perceived severity and beliefs about the causes of obesity

No statistically significant association was identified by three studies (52,53,55) using five univariate analyses exploring the PS–weight association, nor by two studies (52,53) of the BCO–weight association.

Additional considerations in interpreting the studies' findings

Nine of the 12 included studies examined variables other than body weight. These were demographic variables, personal health habits, professional characteristics and recipients' characteristics. However, these variables were tested only once or the effects of these variables appeared conflicting or inconsistent across the studies, except for gender, age, BMI of recipients and relevant knowledge and experience.

Demographic characteristics

Female health professionals were less likely to report negative attitudes to obesity/obese individuals (53,55,56), and this gender difference persisted when BMI and age were controlled (53,56). But other weight management attitude indicators (SE, PRI, OE and PS) did not differ significantly between male and female health professionals. Age was not significantly related to SE and PRI in four studies (47,50,53,56).

Professional characteristics

Relevant clinical experience and knowledge (e.g. weight management training, running a weight loss clinic and being knowledgeable about weight management guidelines) was identified as an independent predictor of SE in two studies (47,52).

Recipients' characteristics

Two quasi-experimental studies examined the influence of recipients' (patients and the general public) BMI on their perceived confidence in receiving advice from health professionals, and both reported no statistically significant associations (48,49).


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

Personal body weight and related attitudes

This review has evaluated personal weight status associated with eight indicators of attitudes to weight management in a variety of health professions.

Not surprisingly, health professionals with normal weight reported higher SE scores, more positive OE of weight management, stronger PRI and fewer PB to weight management practices than overweight health professionals. SE, as the most frequently examined attitude indicator, was weakly, inversely related to personal weight status, i.e. health professionals of normal weight compared with overweight or obese had higher SE, or were perceived as more credible advisors by patients and the public. OE, PRI and PB than SE, showed a stronger association with health professionals' weight status in three of six studies, two of three studies and two of three studies, respectively.

Interestingly, positive BPCOP and ER appeared to be related to being overweight, although its association was weak, which was confirmed by a study of RNs rather than studies of doctors and dietitians. That is, overweight nurses, rather than normal weight nurses, were more likely to hold positive attitudes towards obese patients. A possible explanation is that, health professionals, who struggle to control their weight, may share obese individuals' plight and empathize with their obese patients. This explanation is supported by findings from a qualitative study (57) which found that nurses with a high BMI report that they were able to empathize more strongly with patients who were obese.

Weight status was not found to be significantly associated with the attitude indicators BCO and PS; it seems possible that health professionals, whether overweight or non-overweight, are knowledgeable about the causes and outcomes of obesity, and therefore these attitude indicators are not sensitive discriminators between these two groups of health professionals.

Overall, the findings in this review provide a corrective to the simplistic assumption that overweight health professionals would have entirely negative attitudes towards weight management, as suggested by research into the effect of personal smoking and physical activity on relevant professional attitudes. In fact, this review found weak but complex associations between personal weight status and attitude indicators. These associations may arise mainly from the multi-factorial nature of obesity, which arises from interaction of genetic factors (external uncontrollable factors) and environmental factors (internal controllable factors). Under such circumstances, having a healthy weight (BMI < 25) is no guarantee that the health professional is practising a healthy lifestyle, or has healthy eating and activity habits. For example, there are some health professionals who are of normal weight whose dietary and exercise habits leave much to be desired and who are physically unfit; while some health professionals who are overweight or obese may make healthy lifestyle choices, including making efforts to reduce their own weight. Thus, it is likely that personal weight status is less closely correlated to professional attitudes than specific diet and exercise behaviours. This is illustrated by a study of a random sample of 600 GPs which found that GPs' feelings of effectiveness were associated not only with normal weight (P < 0.05), but with personal success in losing weight (P < 0.01). However, it was not personal weight status but never having dieted themselves (P = 0.05) that was related with GPs' negative attitudes towards obese patients in the multiple logistic regression (47).

Different analysis strategies across the studies included in this review may have influenced the findings. For instance, the extent to which the effects of confounding variables were statistically controlled varied across the studies and this may have resulted in variable findings based on these differences. For example, gender as a potential predictor may mediate the association between personal weight status and relevant attitudes. Also, most studies included were not designed for the purpose of evaluating the association between weight status and related attitudes, and therefore may be underpowered; thus, even if a significant association exists there may have been too few subjects to detect this.

Methodological quality of the included studies

The methodological quality of studies included in this review is moderate when assessed for methodological quality according to criteria identified in reporting guidelines recommended by the EQUATOR Network for different study designs.

There were no prospective studies examining health professionals' weight status associated with attitudes towards weight management. All included studies employed questionnaires as data collection instruments and only one study reported both the reliability and validity of the measure. Moreover, all the studies relied upon self-reported data (e.g. self-reported weight, height, BMI and attitudes towards weight management), which is prone to error and provide weak evidence for addressing the questions studied.

Other limitations included non-random sampling in three cross-sectional surveys, non-random grouping by a quasi-experimental study, low response rates in six cross-sectional surveys with a response rate of <50% and lack of comprehensive analyses – only three studies used multivariable analyses.

Strengths and limitations of the review

To our knowledge, this is the first review to examine the relationship between the weight status of health professionals and attitudes towards weight management. We adopted a wide range of search terms to retrieve all potential articles published in English or Chinese, and strict criteria to include appropriate studies with acceptable methodological quality. As a result, 12 eligible papers were included in the review, which together comprised 14 independent samples of a total of 10 043 respondents. Data synthesis within the review was grounded on an integrative framework of attitudes.

The main limitation of the review is that the included studies were mainly conducted in the USA (eight of 12 studies), and the review was confined to examining some aspects of attitudes (eight attitude indicator) within three types of health professionals (doctors, nurses and dietitians). Thus, it is uncertain that the findings of this review can be extrapolated to other health professional groups. Like many other systematic reviews, this review is limited by the restrictive scope of the literature search and selection procedure. Inclusion of grey literature might have revealed additional studies which should have been included. Limiting the review to studies reported in English or Chinese may have resulted in the non-reporting of studies published in other languages. Finally, the results of our review may be subject to publication bias.

Future research

The relationship between health professionals' weight status and attitudes towards weight management remains understudied. More studies are needed in this field, with particular attention to the following:

  • 1
    Prospective studies are urgently required to better understand the impact of health professionals' weight status on attitudes towards weight management practice.
  • 2
    The construction of a theoretical framework to underpin studies. For instance, drawing upon the theory of reasoned action and the theory of planned behaviour, which have been used frequently in studies of health professionals' intentions and behaviours (46,58), to guide empirical work and thereby lead to greater understanding of the field.
  • 3
    The examination of other important components of attitudes, such as an individual's intention – one of the most immediate predictor of the subsequent behaviour.
  • 4
    The inclusion of a variety of professional samples, such as dietitians, physiotherapists and psychologists/psychiatrists, who have become important members of a multidisciplinary obesity care team; obese patients or the general population, who would present us with other sources of evidence (from recipients' or non-health professionals' views); and samples from Eastern countries, which would allow us to explore the effects of culture and economy.
  • 5
    The employment of appropriately validated instruments, such as well-developed existing attitude scales (e.g. attitudes toward obese persons (ATOP), weight locus of control (WLOC), fat stereotypes questionnaire (FSQ) and implicit association test (IAT)) to improve the accuracy and reliability of data collected and facilitate comparison across the studies.
  • 6
    The use of comprehensive analyses to take more potential factors (e.g. demographics, personal heath habits and professional characteristics) into consideration.

In addition, there is a need for greater transparency in study reporting and closer alignment with reporting guidelines recommended by the EQUATOR Network which would assist in data extraction and synthesis.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References

Data synthesis from the 12 eligible studies indicates that the weight status of health professionals has a weak and complicated association with relevant attitudes towards weight management. Health professionals with normal weight would appear to have higher SE, more positive OE of weight management, stronger PRI and fewer PB to weight management than their overweight counterparts. However, overweight health professionals are more likely to hold more positive attitudes towards obese patients than normal weight health professionals. There was no significant difference in perceptions of the causes and outcomes of obesity across health professionals' weight status. Future work with prospective, theory-based studies, which use appropriately validated instruments, is needed to strengthen this area of research.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of Interest Statement
  9. References
  • 1
    Puhl RM, Heuer CA. The stigma of obesity: a review and update. Obesity 2009; 17: 941964.
  • 2
    Budd GM, Mariotti M, Graff D, Falkenstein K. Health care professionals' attitudes about obesity: an integrative review. Appl Nurs Res 2009; doi: 10.1016/j.apnr.2009.05.001.
  • 3
    van Gerwen M, Franc C, Rosman S, Le Vaillant M, Pelletier-Fleury N. Primary care physicians' knowledge, attitudes, beliefs and practices regarding childhood obesity: a systematic review. Obes Rev 2009; 10: 227236.
  • 4
    Brown I. Nurses' attitudes towards adult patients who are obese: literature review. J Adv Nurs 2006; 53: 221232.
  • 5
    Frank E, Segura C. Health practices of Canadian physicians. Can Fam Physician 2009; 55: 810811, e7.
  • 6
    Brotons C, Bjorkelund C, Bulc M, Ciurana R, Godycki-Cwirko M, Jurgova E, Kloppe P, Lionis C, Mierzecki A, Pineiro R, Pullerits L, Sammut MR, Sheehan M, Tataradze R, Thireos EA, Vuchak J, network E. Prevention and health promotion in clinical practice: the views of general practitioners in Europe. Prev Med 2005; 40: 595601.
  • 7
    Miller SK, Alpert PT, Cross CL. Overweight and obesity in nurses, advanced practice nurses, and nurse educators. J Am Acad Nurse Pract 2008; 20: 259265.
  • 8
    Cross-Government Obesity Unit. Healthy Weight, Healthy Lives: One Year On. Department of Health, England. 2009.
  • 9
    Aboyans V, Pinet P, Lacroix P, Laskar M. Knowledge and management of smoking-cessation strategies among cardiologists in France: a nationwide survey. Arch Cardiovasc Dis 2009; 102: 193199.
  • 10
    Batlle E, Boixet M, Agudo A, Almirall J, Salvador T. Tobacco prevention in hospitals: long-term follow-up of a smoking control programme.[see comment]. Br J Addict 1991; 86: 709717.
  • 11
    Behbehani NN, Hamadeh RR, Macklai NS. Knowledge of and attitudes towards tobacco control among smoking and non-smoking physicians in 2 Gulf Arab states. Saudi Med J 2004; 25: 585591.
  • 12
    Christensen B. [General practitioners and smoking. General practitioners' knowledge, attitudes and smoking habits and the relationship between these assessed by a questionnaire study in the county of Aarhus]. Ugeskr Laeger 1993; 155: 307310.
  • 13
    Dekker HM, Looman CW, Adriaanse HP, van der Maas PJ. Prevalence of smoking in physicians and medical students, and the generation effect in The Netherlands. Soc Sci Med 1993; 36: 817822.
  • 14
    Parna K, Rahu K, Barengo NC, Rahu M, Sandstrom PH, Jormanainen VJ, Myllykangas MT. Comparison of knowledge, attitudes and behaviour regarding smoking among Estonian and Finnish physicians. Soz Praventivmed 2005; 50: 378388.
  • 15
    Yan J, Xiao S, Ouyang D, Jiang D, He C, Yi S. Smoking behavior, knowledge, attitudes and practice among health care providers in Changsha city, China. Nicotine Tob Res 2008; 10: 737744.
  • 16
    Padula CA. Nurses and smoking: review and implications. J Prof Nurs 1992; 8: 120132.
  • 17
    Gunes G, Karaoglu L, Genc MF, Pehlivan E, Egri M. University hospital physicians' attitudes and practices for smoking cessation counseling in Malatya, Turkey. Patient Educ Couns 2005; 56: 147153.
  • 18
    Wells KB, Lewis CE, Leake B, Ware JE, Jr. Do physicians preach what they practice? A study of physicians' health habits and counseling practices. JAMA 1984; 252: 28462848.
  • 19
    Icli F, Icli T, Gunel N, Arikan R. Cigarette smoking among young physicians and their approach to the smoking problem of their patients. J Cancer Educ 1992; 7: 237240.
  • 20
    Notani PN. The role of general medical practitioners in tobacco control programmes: a study in Bombay, India. Health Educ Res 1991; 6: 121124.
  • 21
    Samuels N. Smoking among hospital doctors in Israel and their attitudes regarding anti-smoking legislation. Public Health 1997; 111: 285288.
  • 22
    Becker DM, Myers AH, Sacci M, Weida S, Swank R, Levine DM, Pearson TA. Smoking behavior and attitudes toward smoking among hospital nurses. [see comment]. Am J Public Health 1986; 76: 14491451.
  • 23
    González S, Bennasar M, Pericà J, Segu P, De Pedro J. Spanish primary health care nurses who are smokers: this influence on the therapeutic relationship. Int Nurs Rev 2009; 56: 381386.
  • 24
    Hall S, Vogt F, Marteau TM. A short report: survey of practice nurses' attitudes towards giving smoking cessation advice. Fam Pract 2005; 22: 614616.
  • 25
    Heath J, Andrews J, Kelley FJ, Sorrell J. Caught in the middle: experiences of tobacco-dependent nurse practitioners. J Am Acad Nurse Pract 2004; 16: 396401.
  • 26
    Nardini S. Personal smoking habit and attitude toward smoking among the health staff of a general hospital. Monaldi Arch Chest Dis 1998; 53: 7478.
  • 27
    Kawakami M, Nakamura S, Fumimoto H, Takizawa J, Baba M. Relation between smoking status of physicians and their enthusiasm to offer smoking cessation advice.[see comment]. Intern Med 1997; 36: 162165.
  • 28
    McKenna H, Slater P, McCance T, Bunting B, Spiers A, McElwee G. Qualified nurses' smoking prevalence: their reasons for smoking and desire to quit. J Adv Nurs 2001; 35: 769775.
  • 29
    Nardini S, Bertoletti R, Rastelli V, Donner CF. The influence of personal tobacco smoking on the clinical practice of Italian chest physicians. Eur Respir J 1998; 12: 14501453.
  • 30
    Parna K, Rahu K, Rahu M. Smoking habits and attitudes towards smoking among Estonian physicians. Public Health 2005; 119: 390399.
  • 31
    Perrin PC, Merrill RM, Lindsay GB. Patterns of smoking behavior among physicians in Yerevan, Armenia. BMC Public Health 2006; 6: 139.
  • 32
    Pretti G, Roncarolo F, Bonfanti M, Bettinelli E, Invernizzi G, Ceccarelli M, Carreri V, Tenconi MT. [Survey among GP's about their smoking habits, opinions and behaviours in smoking prevention in Lombardy (Northern Italy)]. Epidemiol Prev 2006; 30: 343347.
  • 33
    Faehnrich J, Gerlach RW. Smoking history and attitudes of nurses at a large tertiary care facility. J Cancer Educ 1989; 4: 5560.
  • 34
    Morra ME, Tish Knobf MK. Comparison of nurses' smoking habits: the 1975 DHEW survey and Connecticut nurses, 1982. Public Health Rep 1983; 98: 553557.
  • 35
    Reeve K, Adams J, Kouzekanani K. The nurse as exemplar: smoking status as a predictor of attitude toward smoking and smoking cessation. [erratum appears in Cancer Pract 1996 Jul–Aug;4(4):223]. Cancer Pract 1996; 4: 3133.
  • 36
    Slater P, McElwee G, Fleming P, McKenna H. Nurses' smoking behaviour related to cessation practice. Nurs Times 2006; 102: 3237.
  • 37
    Al-Doghether M, Al-Tuwijri A, Khan A. Obstacles to preventive intervention. Do physicians' health habits and mind-set towards preventive care play any role? Saudi Med J 2007; 28: 12691274.
  • 38
    Frank E, Bhat Schelbert K, Elon L. Exercise counseling and personal exercise habits of US women physicians. J Am Med Womens Assoc 2003; 58: 178184.
  • 39
    Harsha DM, Saywell RM, Jr, Thygerson S, Panozzo J. Physician factors affecting patient willingness to comply with exercise recommendations. Clin J Sport Med 1996; 6: 112118.
  • 40
    Rogers LQ, Gutin B, Humphries MC, Lemmon CR, Waller JL, Baranowski T, Saunders R. Evaluation of internal medicine residents as exercise role models and associations with self-reported counseling behavior, confidence, and perceived success. Teach Learn Med 2006; 18: 215221.
  • 41
    McKenna J, Naylor PJ, McDowell N. Barriers to physical activity promotion by general practitioners and practice nurses. Br J Sports Med 1998; 32: 242247.
  • 42
    Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process 1991; 50: 179211.
  • 43
    Dunton GF, Kaplan J, Wolch J, Jerrett M, Reynolds KD. Physical environmental correlates of childhood obesity: a systematic review. Obes Rev 2009; 10: 393402.
  • 44
    Ferreira I, van der Horst K, Wendel-Vos W, Kremers S, van Lenthe FJ, Brug J. Environmental correlates of physical activity in youth – a review and update. Obes Rev 2007; 8: 129154.
  • 45
    Michie S, Johnston M, Abraham C, Lawton R, Parker D, Walker A. Making psychological theory useful for implementing evidence based practice: a consensus approach. Qual Saf Health Care 2005; 14: 2633.
  • 46
    Godin G, Belanger-Gravel A, Eccles M, Grimshaw J. Healthcare professionals' intentions and behaviours: a systematic review of studies based on social cognitive theories. Implement Sci 2008; 3: 36.
  • 47
    Bocquier A, Verger P, Basdevant A, Andreotti G, Baretge J, Villani P, Paraponaris A. Overweight and obesity: knowledge, attitudes, and practices of general practitioners in france. Obes Res 2005; 13: 787795.
  • 48
    Hash RB, Munna RK, Vogel RL, Bason JJ. Does physician weight affect perception of health advice? Prev Med 2003; 36: 4144.
  • 49
    Hicks M, McDermott LL, Rouhana N, Schmidt M, Seymour MW, Sullivan T. Nurses' body size and public confidence in ability to provide health education. J Nurs Sch 2008; 40: 349354.
  • 50
    Perrin EM, Flower KB, Ammerman AS. Pediatricians' own weight: self-perception, misclassification, and ease of counseling. Obes Res 2005; 13: 326332.
  • 51
    Watson L, Oberle K, Deutscher D. Development and psychometric testing of the nurses' attitudes toward obesity and obese patients (NATOOPS) scale. Res Nurs Health 2008; 31: 586593.
  • 52
    Hoppe R, Ogden J. Practice nurses' beliefs about obesity and weight related interventions in primary care. Int J Obes Relat Metab Disord 1997; 21: 141146.
  • 53
    Foster GD, Wadden TA, Makris AP, Davidson D, Sanderson RS, Allison DB, Kessler A. Primary care physicians' attitudes about obesity and its treatment. Obes Res 2003; 11: 11681177.
  • 54
    Hendershot C, Telljohann SK, Price JH, Dake JA, Mosca NW. Elementary school nurses' perceptions and practices regarding body mass index measurement in school children. J Sch Nurs 2008; 24: 298309.
  • 55
    Story MT, Neumark-Stzainer DR, Sherwood NE, Holt K, Sofka D, Trowbridge FL, Barlow SE. Management of child and adolescent obesity: attitudes, barriers, skills, and training needs among health care professionals. Pediatrics 2002; 110: 210214.
  • 56
    Warner CH, Warner CM, Morganstein J, Appenzeller GN, Rachal J, Grieger T. Military family physician attitudes toward treating obesity. Mil Med 2008; 173: 978984.
  • 57
    Brown I, Thompson J. Primary care nurses' attitudes, beliefs and own body size in relation to obesity management. J Adv Nurs 2007; 60: 535543.
  • 58
    Perkins MB, Jensen PS, Jaccard J, Gollwitzer P, Oettingen G, Pappadopulos E, Hoagwood KE. Applying theory-driven approaches to understanding and modifying clinicians' behavior: what do we know? Psychiatr Serv 2007; 58: 342348.