Poor nutrition is considered to be the most modifiable risk factor for long-term health(Reference Warber, Warber and Simone1, Reference Kraus and Fox2). Several illnesses such as CVD, type 2 diabetes, hypertension and numerous cancers are linked, at least in part, to poor diet(Reference Kohlmeier3–Reference Nicholas, Roberts and Pond6). The success of treatments and interventions used to address these illnesses hinges on improving diet and nutrition status(Reference Temple7, Reference Schulman8).
Physicians are generally considered by the public as a reliable source of information related to nutrition(Reference Hu, Wu and Liu4, Reference White, Young and Lasswell5). However, a study with physicians in Alberta, Canada revealed that 42 % described their knowledge of nutrition as weak(Reference Temple9). Nutrition knowledge questionnaires of physicians from California(Reference Mlodinow and Barrett-Connor10) and from Canada(Reference Temple9) revealed correct response rates of only 69 % and 63 %, respectively. Primary care physicians in Taiwan(Reference Schulman8) and family practice residents in Texas(Reference Kirby, Chauncey and Jones11) scored even more poorly (59 % and 51 %, respectively). A 1995 national survey of American physicians revealed that many barriers – lack of nutrition knowledge, lack of time, poor patient compliance and inadequate counselling skills – prevented physicians from giving proper dietary counselling(Reference Kushner12).
Identifying weaknesses in nutrition knowledge among Kuwaiti physicians may provide guidance to improve their understanding of nutrition in the future. Therefore, the aim of the present study was to determine whether or not differences exist in the nutrition knowledge of male and female physicians working in Kuwait City, Kuwait.
Methods
A cross-sectional study was conducted on randomly selected male and female Kuwaiti physicians employed by hospitals that are part of the Ministry of Health in Kuwait City, Kuwait. The sample size for the study was calculated on the basis of a 2 % difference in the total percentage score of nutrition knowledge between male and female physicians. From previous studies(Reference Al-Numair13, Reference Al-Zahrani and Al-Raddadi14), an estimated standard deviation of the total percentage score of nutrition knowledge was 2·74 %. The effect size was computed using Cohen's effect size measure (d) for comparing differences in two independent groups. To detect this difference with a power of 80 % at the 5 % significance level (two-sided), each of the two groups (males and females) required thirty-one subjects. To account for non-participation, the sample size was increased to 100 physicians (fifty males; fifty females) working in governmental hospitals in Kuwait City, Kuwait.
A nutrition knowledge questionnaire was adopted and modified from Temple(Reference Temple9). In the first part of the questionnaire respondents were asked to give their gender, age, area of specialty and to self-rate their nutrition knowledge as poor, moderate or excellent(Reference Temple9). The second part of the questionnaire consisted of sixteen multiple-choice questions, each with three possible answers. The questionnaire was pre-tested on eight physicians (four males; four females) by cognitive interviews (concurrent probes) to assess comprehension of the questions. The nutrition knowledge questionnaire was hand-delivered to each of the 100 Kuwaiti physicians working in various medical departments such as primary health care, nephrology, surgery, paediatrics and gynaecology in Kuwait City, Kuwait. The non-respondents received a second questionnaire and a telephone call. The completed questionnaires were received in a sealed envelope. The sixteen questions and the correct answers are presented in Table 1.
Table 1 Nutrition knowledge questionnaire
*indicates the correct answer.
Data were analysed using the SPSS statistical software package version 11·0. Descriptive statistics were used to display data in percentages and mean. Percentages have been rounded-off to the whole value. Student's t test was conducted to assess the differences in nutrition knowledge by gender, age and specialization.
Results
A response rate of 73 % was achieved in the current study (forty males; thirty-three females). The age range of male physicians was 25–57 years, mean 40 years; while the age range of female physicians was 25–53 years, mean 38 years. From the seventy-three respondents, thirty-five were primary health care physicians, nineteen were surgeons, sixteen were paediatricians and three were gynaecologists.
The mean score for correctly answered questions in the overall sample was 60 %, with males and females averaging 56 % and 65 %, respectively (P = 0·042). The percentages of correct responses to the sixteen questions from the questionnaire are presented for the overall sample and according to gender in Table 2. There were no significant differences between genders on most of the questions (P > 0·05) except for questions 2 (P = 0·019) and 8 (P = 0·002).
Table 2 Percentages of correct answers to the nutrition knowledge questions, overall and according to gender, among Kuwaiti physicians (forty males; thirty-three females) working in Kuwait City, Kuwait
*Based on independent-samples t test with df = 71.
Differences in nutrition knowledge scores by age group (<40 years; ≥40 years) are shown in Table 3. Both age groups had an equal mean percentage of correctly answered questions (60 %).
Table 3 Percentages of correct answers to the nutrition knowledge questions, according to age group, among Kuwaiti physicians (forty males; thirty-three females) working in Kuwait City, Kuwait
*Based on independent-samples t test with df = 71.
Table 4 shows that 74 % (41 % of males; 33 % of females) of the physicians described their knowledge of nutrition as ‘moderate’. All physicians who rated their nutrition knowledge as ‘poor’ were in the <40 years age group. The proportion of physicians who rated their nutrition knowledge as ‘poor’ was higher among males (7 %) compared with females (3 %). Furthermore, the proportion of those who rated their nutrition knowledge as ‘excellent’ was higher among males (10 %) than females (6 %). Out of the seventy-three respondents, four females chose not to rate their nutrition knowledge, thereby reducing the number of replies for this question to sixty-nine.
Table 4 Counts of self-rated nutrition knowledge, according to age group and gender, among Kuwaiti physicians (forty males; thirty-three females) working in Kuwait City, Kuwait
Discussion
The response rate in the current study (73 %) was higher than what was reported by Al-Numair (56 %) in Saudi Arabia(Reference Al-Numair13), but higher than that (27 %) in a study of nutrition knowledge among primary care physicians in Taiwan reported by Schulman(Reference Schulman8). Al-Zahrani and Al-Raddadi's(Reference Al-Zahrani and Al-Raddadi14) survey on nutrition knowledge of primary health care physicians in Jeddah, Saudi Arabia showed the same response rate as observed in the present study. In two distinct surveys of nutrition knowledge of physicians, one conducted by Temple(Reference Temple9) in Canada and the other by Mlodinow and Barrett-Connor(Reference Mlodinow and Barrett-Connor10) in California, the response rates were 36 % and 40 %, respectively. These results were comparatively lower than the response rate (73 %) achieved in our study. Seventy-four per cent of the physicians in the present study self-rated their nutrition knowledge as ‘moderate’. This could be one of the reasons for the higher response rate, as physicians might be more curious to know how their nutrition knowledge was.
The mean score for correctly answered questions in the present study (60 %) was slightly lower than that observed in the survey of nutrition knowledge among physicians in Canada (63 %)(Reference Temple9) and comparatively higher than the score (52 %) reported by Al-Numair(Reference Al-Numair13) for physicians working in Saudi Arabia. The mean score for correctly answered questions in the California study (70 %)(Reference Kirby, Chauncey and Jones11) was higher than we found (60 %). However, the questions in the California study were true–false indicating that chance would have raised the score far more than was the case here in our study. Other studies published by Kirby et al.(Reference Kirby, Chauncey and Jones11) and Al-Zahrani and Al-Raddadi(Reference Al-Zahrani and Al-Raddadi14), using multiple-choice questions, also reported low scores of 51 % and 52 %, respectively.
The results of the present study in Table 2 indicate that Kuwaiti physicians were aware of information publicized in the medical press such as the role of n-3 fatty acids, foods and nutrients that are protective against cancer, hypertension, neural tube defects and number of kilocalories in one gram of fat (notably questions 3, 7, 11, 12, 13 and 15). This tendency was similar to findings of Al-Numair(Reference Al-Numair13) and Temple(Reference Temple9). However, a low number of physicians gave the correct answer to questions that are less likely to appear in the press (questions 1, 2, 4, 5, 6, 8, 9, 10, 14 and 16), showing poor knowledge of topics including the role of soluble fibre in lowering blood cholesterol level, the effect of consuming too much protein on body Ca, the adequate intake level of Ca, the major type of fat in olive oil and hydrogenated fats, and the functions and sources of different vitamins and minerals.
Conclusions
Overall, the present results indicate that there are severe gaps in nutrition knowledge among Kuwaiti physicians working in Kuwait City, Kuwait. Several studies along with the current one suggest that many physicians do not have the knowledge to properly advise their patients on the essential role of nutrition in the causation, prevention and treatment of different diseases. Therefore, nutrition should be a fundamental part of continuing medical education and it should be properly integrated into the curriculum of medical schools.
Acknowledgements
This project was supported by Kuwait University. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. A.R.A. wrote the protocol of the study, performed all data analysis and wrote the manuscript. F.A. and A.A.-H. provided significant advice and consultation on the manuscript. The authors are sincerely grateful to Dr Carol Ghazaii for providing valuable assistance and guidance throughout the course of this work.
