Parents’ perceptions and attitudes on childhood obesity: A Q-methodology study

Authors


  • Source of support None.

  • Disclosure
    The authors declare that they have no conflicts of interest.

Correspondence Noori Akhtar-Danesh, PhD, Faculty of Health Sciences, McMaster University, 1200 Main St. West, Room 3N28B, Hamilton, ON L8N 3Z5, Canada. Tel: 905-525-9140; Fax: 905-521-8834; E-mail: daneshn@mcmaster.ca

Abstract

Purpose: The purpose of this study was to investigate parents of young children for their perceptions on the causes of obesity, the impact of childhood obesity on health, and the barriers to successful prevention of childhood obesity.

Data sources: The target population included parents who attended a clinic for their well-baby check-up. The study was conducted in two phases. Using Q-methodology, 33 parents were classified into two groups representing two viewpoints: “confident in delivering healthy nutrition” and “family physical activity focused.”

Conclusion: This work indicates that parents have varying foci on causation of obesity, and differ in focus on nutrition and physical activity. Most of the parents in this study were aware of healthy nutrition, and about one third of them believed in the benefits of physical activity for children and did not see being overweight or obese as a barrier to physical activity. The first group was confident in being able to deliver healthy nutrition to their family, and the second group was characterized by a focus on physical activity and its role in childhood obesity. Both groups agreed that exercising and sports are very important to a child's health status.

Implications for practice: Nurse practitioners have a unique role in the health system and are one of the best facilitators to deliver health messages to the public; thus, they are able to educate parents and increase their awareness about the causes and consequences of childhood obesity.

Childhood obesity is a growing public health concern. Overall, the prevalence of obesity in children aged 2–5 has increased from 5.0% to 13.9% between 1980 and 2003; in the same time, it increased from 6.5% to 18.8% in children aged 6–11 and from 5.0% to 17.4% in youth aged 12–19 (Centers for Disease Control National Center for Health Statistics, 2004). It has been reported that girls who are overweight (body mass index [BMI] >85th percentile) by age 5 are at an increased risk for uninhibited overeating, weight concerns, and body dissatisfaction later on (Shunk & Birch, 2004). These statistics are worrisome given that obese children are at an increased risk of becoming obese adults and developing cardiovascular health problems.

Environmental factors, lifestyle preferences, and cultural environment play pivotal roles in the rising prevalence of obesity worldwide (Dehghan, Akhtar-Danesh, & Merchant, 2005). Identified nutritional risk factors include excessive sugar intake in soft drinks, low fruit and vegetable intake, and increased portion size (Dehghan et al., 2005). Reduced physical activity and increased sedentary time also play major roles in obesity development (Dehghan et al., 2005).

In North America, preschool-aged children spend an increasing amount of time in childcare. The percentage of children aged 3–5 years enrolled in center-based early childhood care and education programs increased from 53% in 1991 to 60% in 1999 (Wirt et al., 2002). Many children eat at least one meal and one snack at daycare. Since children's dietary practices and activity patterns are influenced by what occurs in the childcare environment, it is reasonable to suggest that childcare providers and teachers can act as important mediators in the prevention of childhood obesity. At an older age, the environment also plays a role in the development of childhood obesity through physical activity patterns and food intake at school (Kohl, III & Hobbs, 1998). Young children pursue the behaviors of others in their environment through imitating dietary habits, food preferences, eating patterns, use of foods, and habits of exercise or sedentary activities such as television viewing (Dennison & Boyer, 2004; Edmunds, Waters, & Elliott, 2001). These behavioral determinants and environmental factors need to be taken into consideration when formulating strategies to prevent overweight among children (Goran, 1998). Although there is a general agreement that prevention must be a key strategy for controlling the current epidemic of obesity and that children should be considered the priority population for intervention strategies (Dehghan et al., 2005), there has been limited success to such programs (Hodges, 2003; Myers & Vargas, 2000). On the other hand, family and environmental factors can impact children's weight (Fisher & Birch, 1995). Parents act as agents of food socialization for children (Savage, Fisher, & Birch, 2007). The early childhood years comprise a critical period of development characterized by increases in both cell size and cell number, and overnutrition during this period can set the stage for becoming overweight or obese for a lifetime (Ong, Ahmed, & Dunger, 2006).

inline image

Although behaviors leading to childhood obesity occur within the context of the family where parents are usually responsible for the variety and quality of food available to children (Baughcum, Chamberlin, Deeks, Powers, & Whitaker, 2000), parents' attitudes and perceptions are not often a key part of obesity interventions, rather the interventions target schools and environments.

The aim of this study was to elicit the attitudes and perceptions held by parents of infants and young children up to 3 years old about childhood obesity. Specifically, Q-methodology was used to clarify parents’ perceptions of the causes of childhood obesity, the impact of childhood obesity on health, and the barriers to successful childhood obesity prevention programs. The methodology box summarizes reasons for selection of this approach.

Methods

Q-methodology

Q-methodology was used to identify parents’ common attitudes and perceptions. In this research method, subjective viewpoints are analyzed using a combination of qualitative and quantitative techniques (Akhtar-Danesh, Baumann, & Cordingley, 2008). This method has been used in different areas of health research, including weight control in obese women (Dennis & Goldberg, 1996), evaluation of job satisfaction (Chinnis, Summers, Doerr, Paulson, & Davis, 2001), patients’ viewpoints about health and rehabilitation (Ockander & Timpka, 2005), and clinical decision making (McCaughan, Thompson, Cullum, Sheldon, & Thompson, 2002; Thompson et al., 2001).

Although introduced in 1935 by Stephenson (Stephenson, 1935a, 1935b), Q-methodology has recently been more widely used as a result of advances in the statistical analysis component (McKeown & Thomas, 1988). This method is used to identify diverse viewpoints, as well as commonly shared views, and is particularly useful in research that explores human perceptions and interpersonal relationships (Chinnis et al., 2001).

A Q-methodological study has two phases: (a) development of a sample of statements, Q-sample, related to the topic of interest and (b) rank-ordering of the Q-sample by a group of individual from their points of view, preferences, judgments, or feelings about the statements using a Q-sort table (a grid) with a quasi-normal distribution (e.g., Figure 1). Then, using factor analysis statistics the Q-sorts that correlate significantly with each other are found. Each resultant group of similar Q-sorts represents a group of like-thinking participants. Thus, each factor represents a group of individuals with similar views, feelings, or experiences about the theme of the study.

Figure 1.

A typical Q-sort table for rank-ordering Q-sample statements.

In Q-methodology, a by-person factor analysis (i.e., the statistical analysis is performed by person rather than by variable, trait, or statement) is used to identify different factors, and respondents are grouped based on the similarities of their Q-sorts. One individual is counted on one factor if his/her factor loading is statistically significant (p≤ .05). A factor loading is a correlation between a Q-sort and the factor itself. Then, each identified factor is interpreted based on its distinguishing statements, which define the uniqueness of each factor compared to other factors. In addition, statements with extreme scores on either end of the sorting continuum are of particular interest as they represent the most defining likes and dislikes of the participants loaded on each factor (Valenta & Wigger, 1997).

The test-retest reliability of Q-sorting has been found to be 0.80 or higher (Dennis, 1988, 1992). Content validity is typically assessed by literature review and a team of domain experts and tested in one or more pilot studies. The face validity of the statements is assured by using participants’ exact wording of the statements, if possible, with slight editing only for grammar and readability (Dennis, 1992).

Q-studies typically use small sample sizes, and low response rates do not bias the results because the primary objective is to identify a typology, not to test the typology's proportional distribution within the larger population (Brown, 1993).

Phase 1: Data collection instrument

Identification of statements (referred to as the concourse) After obtaining approval of the Institutional Review Board, we used a convenience sampling approach to survey 20 parents attending a Medical Center in Canada for their well-baby check-up to determine their views and feelings regarding childhood obesity. A trained research assistant described the study to the parents and asked them to answer a preliminary qualitative open-ended questionnaire assessing their understanding of good health, healthy and unhealthy food, their current eating behavior, long-term risks of unhealthy nutrition, childhood obesity, talking with other parents about children's nutrition and its relationship to health, exercising and sports, their current exercising behavior or physical activity level, ideal weight, and barriers to physical activity in children, specifically in obese children. Responses to the questionnaire were summarized into different statements.

Compiling the Q-sample statements Members of the research team independently reviewed all the statements from the original 20 participants (using respondents own words whenever possible) for similarities and differences. Duplicates were discarded. The research team then reached consensus on a set of 42 statements (henceforth Q-sample), which covered all the major views presented in the statements. These statements were then numbered randomly and each was typed onto a card.

A Q-sort table (a grid) was then developed, which involved a quasi-normal distribution containing 42 cells equal to the number of statements in the Q-sample (Akhtar-Danesh et al., 2008). An instruction page was prepared for Q-sorting of the statements. A convenience sample of five parents were asked to pilot test this data collection instrument and provide their feedback on clarity, ease of the task, length of time for completion, and general information about the process. Their suggestions were used for further modifications of the instrument and the instructions.

Phase 2: Data collection and analysis

In this stage of the study, a convenience sample of 100 parents from the same clinic were contacted by telephone and invited to participate in the study to rank order the 42 statements from the Q-sample. A package containing the Q-sample, Q-sort table, instructions, a short demographic questionnaire, and two copies of the consent form was mailed to each parent who agreed to participate in the study. In order to increase participation, 2 weeks later a reminder letter was mailed to all nonrespondent parents. The instruction letter to participants requested that they complete the questionnaire and then rank order the statements along a continuum from “most disagree” at one end to “most agree” at the other end using the Q-sort table in a two-step process. They were instructed first to read all statements to get an impression of the range of opinion and sort the statements by initially dividing them into three piles as agreeable, disagreeable, and neutral. Second, they were asked to rank-order the statements from least agree to most agree. Participants then mailed back the completed Q-sort tables and questionnaire along with the signed consent form to the principal investigator.

Data analysis

The rank-ordered scores were entered into the PQMethod 2.11 program (Schmolck, 2002) for analysis and were analyzed using a by-person factor analysis technique to form groups (factors) of parents based on the similarities of their Q-sorts.

Results

Overall 33 parents, of which 32 were mothers and one was a father, completed the Q-sort table and the demographic questionnaire. One parent questionnaire was excluded from the analysis due to an incomplete Q-sort table. Mean age for the parents and their children was 34.4 (SD= 4.3) years and 16.0 (SD= 7.8) months, respectively. Based on self-reported weight and height of the 33 parents, 25 (75.8%) were normal weight (BMI 18–24), seven were overweight (BMI 25–29), and one was obese (BMI ≥ 30). The participating parents had college/university level education (Table 1). Using a by-person factor analysis, two factors emerged representing two salient viewpoints. Each factor was named based on its distinguishing statements. A summary of the distinguishing statements is presented in Table 2. No statistically significant difference in the parent and child age, gender of child, parental BMI, education, or income level was identified between factors. One parent did not load significantly on either factor.

Table 1.  Some descriptive statistics of parents and children
 MeanSDa
  1. aStandard deviation.

  2. b2 missing.

Parent's age (year) (n= 33)34.44.3
Child's age (month) (n= 32)16.07.8
Parent's BMI (n= 33)23.64.6
N%
Child's gender
 Male2266.7
 Female1133.3
Parent's BMI group
 Normal2575.8
 Overweight721.2
 Obese13.0
Parent's number of childrenb
 12064.5
 2825.8
 339.7
Table 2.  Distinguishing statements for the factors
Stat no.StatementFactor 1Factor 2
  1. Note. Score changes from 4 to +4 and negative score indicate disagreement.

Factor no. 1: Confident in delivering healthy nutrition
 17I eat fruit and vegetables at least two to three times a day.31
 20Unhealthy foods are more convenient and cheaper.−31
 37My eating behavior is good for a few days a week and bad for a few days a week.−31
 24I have difficulty managing my busy lifestyle and I tend to rely on quick and easy foods.−32
 2I think it is difficult to have my family consume a healthy, balanced diet and to engage in frequent physical activities.−41
 26I almost never eat breakfast.−43
Factor no. 2: Family physical activity focused
 16I think that everything in moderation is okay. Fresh foods (fruits, veggies, meats, and dairy) are healthy. I try to avoid fast food and processed food when possible.13
 15I think one important barrier to physical activity in obese children is that it is more difficult to participate in activities because of the obesity itself.0−2
 12I think that an important barrier to physical activity in children is that there is little opportunity for physical activity in schools.1−3
 29I think that walking 30 min each day is too little activity for good health.1−3
 13It is better not to talk with other parents about children's nutrition as you do not want to cause trouble over a clash of values.2−4
 30It is unhealthy to include fats in my child's diet.2−4

Factor no.1: Confident in delivering healthy nutrition

Nineteen parents loaded significantly on this factor. This group believed that they were delivering healthy nutrition to themselves and their family and saw few “excuses” for not doing so. They consumed fruit and vegetables at least two to three times a day, disagreed that unhealthy foods were more convenient and cheaper, expressed that they did not have difficulty managing their busy lifestyle, did not rely on quick and easy foods, and strongly disagreed that they almost never eat breakfast or that their eating behavior was good for a few days a week and bad for a few days a week. They disagreed with the statement that “I think it is difficult to have my family consume a healthy, balanced diet and to engage in frequent physical activities.” This group was more neutral in their thoughts on physical activity.

Factor no.2: Family physical activity focused

Out of 33 parents, 13 of them loaded on the second factor. This group was characterized by a focus on physical activity and its role in childhood obesity. They disagreed that there were insufficient physical activity opportunities at school and that being obese was an important barrier to physical activity in children. They did not think that walking 30 min each day was too little activity for good health. They supported the idea that everything in moderation was acceptable. This group believed that fresh foods (fruits, vegetables, meats, and dairy) are healthy and they try to avoid fast food and processed food when possible. They strongly opposed the idea that “It is unhealthy to include fats in my child's diet.” They disagreed that “It's better not to talk with other parents about children's nutrition as you don't want to cause trouble over a clash of values.”

Consensus statements

There were several statements that both groups equally agreed or disagreed with (Table 3). Both groups strongly agreed with the statements that “I think that exercising and sports are very important to a child's health status,”“I think eating healthy foods such as fruits, vegetables and whole grains are important to consume on a regular basis,”“I think that the long-term risks of unhealthy nutrition are heart disease, diabetes, cancer, and poor bone health,” and “I think that processed foods, packaged foods, fast foods, chips, pops, sugared cereal, sweets and chocolate are unhealthy.” Both groups strongly disagreed that “I think it is okay to eat fast foods once a week,”“I think that the most important barrier to physical activity in children is safety concerns re: letting kids outside unsupervised,”“I have sufficient physical activity by doing house work and playing with my children,” and “I wish I could afford more healthy foods.”

Table 3.  Consensus statements
Stat no.StatementFactor 1Factor 2
22I think that exercising and sports are very important to a child's health status.44
21I think eating healthy foods such as fruits, vegetables, and whole grains are important to consume on a regular basis.44
7I think that the long-term risks of unhealthy nutrition are heart disease, diabetes, cancer, and poor bone health.33
28I think that processed foods, packaged foods, fast foods, chips, pop, sugared cereal, sweets, and chocolate are unhealthy.33
9I think that a healthy child is active and energetic.32
27I think that the most important barrier to physical activity in children is lack of adult involvement in helping them in more active play choices.23
10I think that the most important barrier to physical activity in children is television, video games, and the computer.22
36I think that homemade foods are healthier.22
8Dairy products are healthy food.21
42I think that the most important barrier to physical activity in children is having inactive parents lack of adult involvement in helping them in more active play choices.22
18I think it is okay to eat fast foods once a week.−2−2
39I think that the most important barrier to physical activity in children is safety concerns re: letting kids outside unsupervised.−2−2
11I have sufficient physical activity by doing house work and playing with my children.−2−3
35I wish I could afford more healthy foods.−3−2

Discussion

This study highlighted some important issues regarding parent's awareness on childhood obesity. The findings indicated that most of the parents in this study were aware of healthy nutrition and about one third of them believed in the benefits of physical activity for children and did not see being overweight or obese as a barrier to physical activity. Factor 1 demonstrated that parents were confident in their ability to deliver healthy nutrition to their children. Factor 2 focused on issues regarding physical activity rather than on food and eating issues. Although the groups differed in their focus on childhood obesity, there were no differences between the groups in age, BMI, or education. Both groups agreed that exercising and sports are very important to a child's health status. These findings are in agreement with findings by Nazario Rodriguez, Figueroa, Rosado, & Parrilla (2008) that 67% of parents were concerned about their children being overweight or obese and 90% expressed their concern about health consequence of obesity. Similar results were reported by Myers and Vargas (2000), who identified 78% of parents as concerned about the consequences of obesity. Parent awareness has been suggested as an important factor to combat obesity by the Robert Wood Johnson Foundation, which vowed to devote $500 million in funds to combat childhood obesity and reverse the obesity epidemic in American children by 2015 (http://www.rwjf.org/childhoodobesity/). In spite of parents’ awareness shown by many studies, the prevalence of childhood obesity continues to increase, indicating that individual level of knowledge and parental action are not enough to deal with the problem.

Interestingly, in spite of what is known about the probable influence of community/environment on childhood obesity, these parents did not believe environmental characteristics influenced them, at least in the behaviors they identified with; that is, Factor 1 was confident in nutrition, Factor 2 was confident in physical activity capabilities. Perhaps, identifying parental confidence in knowledge of healthy behaviors may be important for successfully implementing lifestyle behavior changes, and thus be important modulators of the success of prevention programs.

Although parents may provide the strongest influence on children's health beliefs and behaviors, they are not the only ones influencing their behaviors. At school, peers are considered to be particularly influential in adolescent eating behavior. Birch (1980) studied preschool children's behaviors and showed that when children observe other children choosing and eating vegetables that they did not like, their preferences and intake for disliked vegetables increased. Also, Feunekes, de Meyboom, and van Staveren (1998) assessed adolescents’ habitual food intake using a food frequency questionnaire and found that 19% of foods consumed by adolescents were similar to those consumed by their friends; more specifically, associations with peer intake were found for snack foods. To recognize the causes of childhood obesity, the starting point may be to look at the parents, but this needs to be matched with population-wide and public health activities. The encouragement of an obesogenic environment through media is an aspect of childhood obesity that parents cannot be expected to offset. Coon and Tucker (2002) showed that fast foods or foods high in sugar are heavily advertised during children's television programming.

This study indicated that parents did not view accessibility to unhealthy foods as an important factor contributing to obesity. This may seem contrary to some recent research, which has shown that the environment has a large influence on food intake and energy expenditure (Timperio, Salmon, Telford, & Crawford, 2005). Contradictory results illustrate that parents may not be fully aware of energy density of pre-prepared foods and foods that are available at most fast food restaurants. Literature shows that the culture of eating out is increasing among families (Kant & Graubard, 2004). A survey of foods sold in popular take-out restaurants showed that all food portions except sliced white bread exceeded United States Department of Agriculture (USDA) and Food and Drug Administration (FDA) recommendations (Young & Nestle, 2002). Also, increasing consumption of processed foods, convenience and pre-prepared foods, decline in cooking, and increasing variety of food are contributing factors in increasing the prevalence of obesity in developed and developing world (Banwell, Hinde, Dixon, & Sibthorpe, 2005; Wang, Cubbin, Ahn, & Winkleby, 2008).

Although parents in this study did not find environmental safety as a barrier for physical activity, safety of environment has been identified as an important issue for children's physical activity. A study from Australia found that a large number of school children dropped walking or cycling to school due to safety fears (Australian Institute of Health and Welfare, 2004). We believe that among other environmental factors, providing security and safety for children playing outside the home, safe and practical walking sidewalks to and from schools and shopping areas, providing enough play-grounds, and safe sidewalks in general can be very effective to combat childhood obesity.

Parents are one of the most influential factors in laying the foundation for early childhood weight problems (Doolen, Alpert, & Miller, 2009). Hence, it is important that they have a solid understanding of healthy active behaviors. If parents do not perceive their children as overweight, no intervention will occur. However, Genovesi et al. (2005) argued that it is possible that parents do, indeed, recognize obesity in general but hesitate to label their own children as overweight or obese at a young age, even in an anonymous survey where they may consciously or unconsciously underestimate their child's weight. Parents may hold distorted perceptions unique to their child, while still being capable of recognizing overweight in other children (Jain et al., 2001). It is also possible that parents are privately aware of their child's weight problem, but reluctant to acknowledge it, as some of them may think that having an overweight or obese child suggests that they are a bad parent (Jain et al., 2001). As parents follow the daily information provided by mainstream media about obesity, this may suggest that more up-to-date and detailed information can be easily delivered by policy and decision makers through the media. In addition, some suggestions focus on school-based interventions, including integrated nutritional and physical activity such as noncompetitive sports (e.g., dancing) in the school curriculum and increasing time for physical activity during the school day (Flores, 1995).

Limitations

This study has some limitations. The participants were homogenous in their education. Although education may have played a part in the findings, based on the results extracted from Canadian Community Health Survey dataset by Statistics Canada (Statistics Canada, 2007) about 80% of women in Canada in this age group have at least a post secondary education. Besides, as has been clarified in the method section in a Q-study, it is not the proportion of the participants that is important but their viewpoints. As a result, the findings from a Q-study are not usually generalizable to the larger populations. Also, this study was conducted in a relatively small community where environmental safety may not be an important issue where it is an important issue in larger communities.

Implications for nursing practice

To combat childhood obesity, there is an urgent need for public health interventions as well as educating parents regarding childhood obesity and its health consequences. This study revealed parents’ awareness on childhood obesity, in particular on lifestyle behaviors such as nutrition and physical activity. This work also suggests that parents have varying focus on causation of obesity, and differ in focus on nutrition vis-a-vis physical activity. Evidence suggests that to alter the population energy balance, focus should be on both, but this work suggests parents may be more receptive to hearing messages relating to one behavior or another, based on their perceptions of causation. Future work should focus on applicability of these findings to other populations and examining the potential influence these beliefs or perceptions may have on actual behavior patterns.

Nurse practitioners have a unique role in the health system and are one of the best facilitators to deliver health messages to the public; thus, they are able to educate parents and increase their awareness about the causes and consequences of childhood obesity. They are able to explain not only the issues of nutrition and physical activity but also other important issues of obesogenic environment in general and emphasize to the parents’“under-estimation of overweight and obesity on their own children,” frequency of eating out, processed food intake and school-related physical activities. This awareness, when established in the community, can be transformed into empowerment and practical actions to combat childhood obesity more effectively.

Summary

The findings from this study indicated that most of the parents were aware of healthy nutrition and benefits of physical activity for their children and did not see being overweight or obese as a barrier for physical activity. Most of them were quite confident in their ability to deliver healthy nutrition to their children. Although it is important to further educate parents regarding childhood obesity and its health consequences, there is urgent need for multicomponent commercial and social activities including integrated nutritional and physical activity such as noncompetitive sports (e.g., dancing) in the school curriculum and increasing time for physical activity.

Acknowledgments

We would like to thank all individuals and parents who helped us on this project, including Dr Maryam Rostami, Dr Nahid Faroughi, and Miss Negar Ghoraishi.

Ancillary