Kearney, Margaret H. PhD, RNC; Rosal, Milagros C. PhD; Ockene, Judith K. PhD, and; Churchill, Linda C. MS
FFQ = Food Frequency Questionnaire;, SORC = Stimulus-Organism-Response-Consequence;, WHI = Women’s Health Initiative.
Obesity reaches its highest prevalence in individuals who are between the ages of 53 and 71 years (1), but the predictors of change in dietary intake in older adults are not well known. When dietary fat intake has been reduced, it has returned nearly to baseline by 10 years after intervention (2, 3). The Women’s Health Initiative (4), a 15-year multisite study of postmenopausal women’s health that includes a dietary modification arm, offered a unique opportunity to study factors that affect adherence to dietary change over time. In an exploratory study, data collected from WHI participants in focus groups and telephone interviews were analyzed to better understand influences on adherence or nonadherence to dietary change.
The SORC model of health behavior (5–7) guided the study. Stimulus conditions consisted of external influences, such as presence of a high-fat food or a negative comment from a family member, that act as external facilitators or barriers to the behavior of interest. Organismic factors were defined as conditions within the individual, such as the cognitive factors of beliefs, preferences, self-talk and internal dialogue, affective and physiologic states, and behavioral habits and skills that prompt or constrain the target behavior. Response repertoire referred to the specific actions involved in fat intake. Consequences were positive or negative internal or external events contingent on the target behavior (eg, sensations of hunger or satiety and praise or criticism from others) that in turn influence the continuation or cessation of the behavior.
Previous qualitative research offers preliminary data about external and organismic influences on dietary change as identified in the SORC model. External influences in mainly young to middle-aged subjects have included advertising and other mass media (8, 9), food costs, and lack of time (9–12). Family influences on middle-aged and older women’s eating behavior have included controlling husbands (13) and responsibility to provide satisfying and tasty food to their families (8–10). Organismic (intrapersonal) influences on dietary behavior in past research have included physical influences, including onset of disease or disability in middle age (8) and physical impediments such as dental problems and digestive difficulties in older adults (10). Beliefs that diet is important to health and that health problems can be reduced by dietary change were common influences across age groups (8–11). Emotional prompts to overeat were described by young and middle-aged women who successfully lost weight (14, 15) as well as those who compulsively overate or were obese (16, 17). Cognitive strategies of successful young and middle-aged weight losers included coming to terms with being overweight and making a decision to lose weight (13), changing perspective (14), and using mental self-monitoring (15). In young to middle-aged samples, negative self-image and a desire to lose weight and improve appearance were strong motivators (9, 12–15). However, in one of the few available age comparisons, when adults over 55 were compared with younger respondents in a large health behavior survey (18), older adults were less likely than younger respondents to describe efforts to change their eating behavior.
Little has been reported about the behavioral responses involved in dietary change, the third component of the SORC model. In one study of middle-aged women, dietary fat reduction behaviors included decreasing use of fats as flavorings and in cooking, eating fewer snack foods and desserts, replacing meat with other foods, altering breakfast patterns, and increasing fat-modified foods (19). The fourth component of the model, reinforcing or deterrent consequences of reducing dietary fat, has not been described. The purpose of the present exploratory study was to elicit the experiences of a small group of women over 50 years who were voluntary participants in the WHI dietary intervention to explore influences on and strategies for dietary change as well as the consequences that contributed to or deterred continued adherence to a low-fat diet. This qualitative information was collected to guide instrument development for use in the design of interventions to support older women’s dietary adherence.
Experiential data regarding influences on dietary fat reduction were collected in a small sample of women enrolled in the WHI dietary intervention, in which the goal was to maintain fat intake ≤20% of total calories. Focus groups and telephone interviews were used to collect qualitative data from adherent and nonadherent participants.
All participants had completed (with varying levels of participation) the intensive first year of the WHI dietary modification intervention protocol (20) and were in the maintenance phase that involved quarterly group meetings with a nutritionist, periodic but not continuous food diaries, and voluntary peer group activity participation. To achieve ethnic diversity, women in the present study were drawn from two WHI centers. The Massachusetts center was the recruitment source for white women, and the New York center was used to recruit Hispanic and black women.
Two sources of data were used at both sites to classify women as adherent or nonadherent for the purpose of the study: 1) nutritionists were asked to list their most and least adherent WHI participants based on the women’s food diaries (as part of the intervention, women were asked to keep and later turn in three food diaries per month) and 2) adherence or nonadherence status of the women listed by nutritionists was checked against the women’s Food Frequency Questionnaire scores, which indicate percentage of fat within total calories. To be invited to participate in the study, women had to be classified as adherent or nonadherent by both criteria.
Recruitment was accomplished by mailed letters of invitation to all eligible women, followed by telephone calls. A total of 109 eligible women were sent a letter inviting them to participate in the study, 103 were reached by telephone to follow up on the invitation, and 92 participated, which represents a response rate of 84% of the original total and 89% of those who were called (92% of the white women and 100% of the minority women who were phoned participated). Women were told that the purpose of the study was to identify factors that helped or hindered adherence to the low-fat dietary protocol. Women were not informed of their adherence status grouping. The average fat intake as percentage of total calories was 19% for adherent group (range = 12%–24%) and 32% for the nonadherent group (range = 25%–44%). Among the 49 white participants, all the women classified as adherent had <20% of total calories from fat, and all the nonadherent had >25% of total calories from fat, with the median dietary fat percentage of nonadherent women in the low 30s. Of the 43 Hispanic and black participants, seven of the 16 women determined to be adherent had fat percentages between 21 and 24%. A risk of misclassification was present for these seven women, but a demarcation was apparent between their FFQ scores and those of minority women deemed nonadherent, whose median dietary fat percentage was in the high 30s. The 92 participants, described in Table 1, had an average age of 64 years (range = 55–80). The majority was or had been married, had some post-high school education, and was employed at least part-time.
A combination of focus groups and individual telephone interviews were used. 1 To begin the data collection, 20 white women from Massachusetts participated in focus groups. Four focus groups were conducted by the principal investigator of the study (M.R.) with 20 white women in central Massachusetts. The participants were assigned to homogeneous groups based on age and adherence status, each with three to six women: adherent women over and under 65 years and nonadherent women over and under 65 years. Focus groups were conducted over a period of an hour or longer after a meal or snack of low-fat foods was served. Sessions were audiotaped and transcribed verbatim. Individual participants were identified on the transcripts, enabling frequency of responses to be tabulated.
Telephone interviews were conducted individually by a Massachusetts WHI staff member with 43 adherent and nonadherent black and Hispanic women from the New York City area. Interviews were conducted in English. Language barrier was identified as a problem in one telephone interview, and these responses were not analyzed. Telephone interviews were conducted by another interviewer with 29 white women in Massachusetts. Both interviewers had extensive experience in conducting telephone surveys and were unaware of the participants’ adherence status. A structured script was used by the interviewers, and data were recorded on paper-and-pencil data collection forms. Responses were recorded verbatim whenever possible and were otherwise paraphrased. Consent was obtained before starting each group or telephone interview. Women were informed about the maintenance of confidentiality and were reminded of the voluntary nature of their participation. Institutional review board approval for the study was in place.
Focus groups and telephone interviews used the same protocol of questions and probes based on the SORC model. Questions included reasons for joining the dietary modification trial, perceived benefits of a low-fat diet, personal and environmental triggers for eating high-fat foods, factors that enhanced adherence, and negative and positive consequences experienced by the women contingent on their adherence to a low-fat diet. Social, personal, emotional, and physical experiences were elicited. In an effort to encourage focus group participants to think of and discuss their own experiences and thus minimize the “me too” tendency and the tendency to refrain from contrasting responses, participants were asked before each discussion to fill out a brief sentence completion questionnaire inquiring about external and organismic influences and consequences. (Because this problem was not expected to occur with women who participated in the telephone interviews, these participants were not asked to complete this questionnaire.) No suggestions were made on the questionnaires as to possible or typical responses. 2
Qualitative content analysis was used to analyze focus group and telephone interview data. The content analysis method can entail both quantitative and qualitative strategies. In the first, the frequency of types of data is determined, and meaning is derived from these counts. In the second, the data in each category are examined for qualities, relationships to other categories, and implications (21). In this study, both qualitative and quantitative content analysis of transcribed focus group data and interviewer notes taken during telephone interviews were conducted using commonly accepted techniques (22–24). The responses were coded into latent or inferred categories (25) within the predetermined components of the SORC model. All development of categories and categorization of responses were blinded to adherence status. Subsequently, adherence status was unblinded, and data from each adherence group were then sorted into the major and minor categories using a detailed grid. Paraphrased and verbatim data from telephone interviews were treated equally in analysis.
Frequency counts were used to describe the proportion of participants mentioning each concept. Cross-group differences were thus evaluated both qualitatively (content, similarity, and intensity of elicited responses) and quantitatively (frequency counts of response types within adherence groups). For the purpose of this exploratory study, differences of greater than 10% between proportions of adherent and nonadherent women’s responses (a difference of more than 4 adherent or 5 nonadherent women) were noted as of clinical interest. Test statistics were not calculated because of the qualitative paradigm of the study and the spontaneous (rather than forced-choice) nature of the comments. Although the focus groups were stratified by age (55–65 and over 65 years) and telephone interviewees’ ages were known, no age-related differences were detected; therefore, age groups were not differentiated for reporting of the results.
Reliability (stability) was enhanced by coding all data twice. A single investigator conducted all the original coding, and representative data classifications in all categories were reviewed by a second investigator to establish objectivity and system (22). Validity (generality) of codes was supported by independent elicitation of very similar responses across all group and individual interviews and by coherence of verbatim content samples within and across categories.
External Stimuli or Influences for Dietary Behavior
In discussion of external stimuli for dietary change, no differences were seen between adherent and nonadherent women (Table 2). Sixty percent of adherent women and 62% of nonadherent women described the Women’s Health Initiative dietary intervention program structure—especially the frequent meetings, group support, and monitoring in the early months—as beneficial. Some women said that they needed more intervention and structure; one woman said that she preferred Weight Watchers because they had weekly meetings, placed more focus on weight loss, and had fewer limitations on the types of foods that could be consumed. In contrast, some found the structure of meetings excessive.
Both adherent (55%) and nonadherent women (58%) reported explicit external pressure to participate in social eating and to try foods prepared for them by others. Only 20% of adherent and 15% of nonadherent women reported receiving overt social support for dietary change. A nonadherent Hispanic woman said that others coaxed her to enjoy life: “eat and die happy.” Family meal routines and feeding children and grandchildren were challenges, as were holidays, family celebrations, and other traditional occasions where high-fat foods were served. A nonadherent white woman reported, “Now before the holidays [my friends are] already saying, ‘You going to make those [French Canadian] meat pies?’” Externally determined obligations and events were cited as obstacles by 13% of adherent and 21% of nonadherent participants. Custodial responsibilities such as taking grandchildren to appointments and supervising them interfered with attendance at WHI program meetings.
Half of both groups of women (48% of adherent and 52% of nonadherent) found the presence of high-fat foods or lack of available low-fat options to be barriers to adherence, as an adherent white woman reported: “At work when people bring in doughnuts, muffins, and whatnot and candy. . .and it sits there in front of me. . .you just have to keep reminding yourself how many grams of fat [are in] one piece of candy.”
Organismic (Intrapersonal) Influences
The main differences in adherent and nonadherent women’s responses were seen in organismic influences on dietary behavior. These intrapersonal influences were represented in differing perceptions of themselves and different interpretations of the external influences both groups had reported with equal frequency (Table 3). Adherent women’s pride, commitment to lifelong change, and willingness to stand up to social pressure contrasted with nonadherent women’s perception of social disapproval and loss of important satisfaction when attempting to reduce their fat intake.
Self-perception as able to achieve dietary change.
Despite sharing many positive views of the WHI program as an external influence, adherent and nonadherent women’s internalization of program goals varied. Most women in both groups (63% of adherent and 69% of nonadherent women) described health as an important value and a reason for enrolling in the WHI program, and roughly equal proportions (40% and 31%) viewed themselves as needing support to accomplish their dietary goals. However, whereas 33% of adherent women were able to internalize their program participation and lifelong dietary changes as part of their identities, only one nonadherent woman (2%) described doing so.
Twenty percent of adherent women but only 6% of nonadherent women saw themselves as assertive in making dietary choices and willing to try new things, and 53% of adherent women but only 27% of nonadherent women reported that they had willpower. Nonadherent women saw themselves as having more than average cravings or personal barriers to adherence and admitted that they were less motivated at times. A nonadherent black woman said that she had developed more self-discipline in the WHI program, but “who wants to do that all the time?”
Knowledge and experience.
More adherent women than nonadherent (53% vs. 25%) felt that they now had adequate knowledge to eat well and cited examples of using what they had learned (50% vs. 19%). Twenty-three percent of nonadherent women but no adherent women said that they needed more knowledge to be successful, and indeed several nonadherent women made erroneous statements about fat content of foods. Comments indicated that nonadherent women saw their learning as increasing their restrictions and viewed their food options as fewer than before, whereas adherent women reported now having a greater variety of satisfying food options. They described creatively extending their learning in selecting and preparing foods, using reference materials supplied by WHI.
Internalized routines, habits, and emotional responses.
A majority (62%) of nonadherent women and fewer (38%) adherent women were still influenced in their present shopping and eating patterns by their long-standing internalized food routines and preferences, including preferences for butter, desserts, and traditional ethnic foods. Nonadherent women more often shopped and cooked based on whims rather than planning, as when a nonadherent white woman reported: “I never know what we’re gonna eat, I just—this morning, you know, ‘Okay, we’re gonna have eggs.’” More nonadherent (38%) than adherent women (13%) described feeling unable to change their routines with work, family, or friends to facilitate dietary adherence. Nonadherent women (50%) were more likely than adherent women (33%) to eat their favorite foods to excess, and nonadherent women were more likely (44% vs. 25%) to feel that there were no satisfactory alternatives to high-fat foods when under emotional strain. A nonadherent black woman described how at a stressful time, “an O’Henry [candy bar] was better than a prayer.”
Perceived opinions and expectations of others.
As noted earlier, reported overt expressions of support or demands from others did not differ between adherent and nonadherent women. However, more nonadherent women (62% vs. 35%) were unwilling to risk social disapproval. They described feeling hurt by negative comments and being unwilling to be different from or disappoint others or to ask for accommodations in restaurants or at social occasions. Comments included, “You don’t want to punish your family, after all, they’re not on this diet,” and “You don’t want to hurt their feelings.”
Self-talk and cognitive strategies.
Negative self-talk to rationalize overeating high-fat foods was used by 46% of nonadherent and 18% of adherent women. In contrast, positive self-talk to promote adherence was used by 58% of adherent but only 21% of nonadherent women. Examples of positive self-talk included reminding themselves to be aware of what they were eating and to remember the lifelong nature of their commitment. An adherent black woman said she had reframed fat as “the enemy,” and another told herself, “It’s not worth it for a 10-minute thrill.” Thirty percent of adherent women vs. 15% of nonadherent women remarked that dietary change was their voluntary choice. An adherent woman commented, “There’s nobody standing over you making you do it.”
Response Repertoire: Dietary Fat Consumption Behaviors
As seen in Table 4, more adherent women than nonadherent (45% vs. 33%) had changed cooking methods and types of foods eaten to reduce fat content. Fairly equal proportions of adherent and nonadherent women (20% and 15%) ate selected high-fat foods such as meat and butter less often than before and described restricting their fat intake after overeating (35% and 29%), and many women had reduced their portions of meat and butter, but adherent women were more likely to go to great lengths to try a variety of foods (45% vs. 27%) and ask for alternatives in restaurants (25% vs. 8%).
Both adherent and nonadherent women described efforts to save up and budget fat grams over the course of a day and balance low- and high-fat meals over time, but adherent women were more likely to cut back on high-fat food in anticipation of a social event where high-fat food would be served (28% vs. 12%) and to consume smaller portions of high-fat foods (58% vs. 40%).
Consequences of Dietary Change
Adherent women were more likely than nonadherent women (20% vs. 8%) to report increased energy and well-being with reduced dietary fat intake, as shown in Table 5. More adherent women (48% vs.19%) reported that their tastes had changed and that fat now produced indigestion or unpleasant sensations. A white woman said, “I can feel the grease in my mouth.” Nonadherent women more often reported negative physical sensations with reduced fat intake (29% vs. 13%), such as feeling weak or dizzy, having dry skin and hair, or feeling less healthy overall. Physical satiety was important to nonadherent women. One said she did not feel she had a “full meal” without dessert. Many nonadherent women (44% vs. 5% of adherent women) reported that they felt strong positive physical sensations when they ate high-fat treats. Descriptions like “bliss,” “tastes so good,” “fantastic,” and “delightful” were used. They had experienced fewer rewards for adherence: 37% (vs. 10% of adherent women) could cite no negative consequences of eating fat, and 27% (vs. 10%) could cite no positive consequences of refusing it.
Psychological and social consequences.
Pride in their successful adaptation to the program and satisfaction with adherence were more common for adherent women (28% vs. 13%). These adherent women described a strong positive change in self-image, and some women mentioned a spillover of increased confidence into other areas, whereas more nonadherent women (35% vs. 20%) described feeling deprived or left out when denied their preferred foods.
Thirty percent of adherent women but only 12% of nonadherent women described changes in their social patterns as a consequence of their low-fat eating. For example, adherent women offered to host social events to ensure that low-fat options would be available and/or to show off their low-fat cooking. They were ready to share their newfound healthy lifestyle. As one woman explained, her friends “aren’t very bright [if] they haven’t figured out I’m real committed to this.”
The overarching contributors to adherence to dietary modification in this sample of healthy older women were making a long-term personal commitment to change and being willing to stand up to others’ disapproval and to replace the satisfaction of eating fat with other kinds of rewards. These internal, organismic strengths were used to overcome external barriers and negative consequences. Nonadherent women were aware of fewer positive effects of dietary change or negative effects of overeating and thus lacked these reinforcements of healthy dietary behavior. Although they reported no more overt external social pressure than did adherent women, their greater intrapersonal concern for social acceptability led to unwillingness to make changes in their social interactions around food. Furthermore, the reinforcement for fat consumption in the tangible satisfaction and satiety that in their view, only high-fat foods could bring was stronger than their desire for the intangible goal of potentially reducing health risks.
The external and organismic influences on dietary behavior in these women resemble those of younger groups in previous studies (8, 9, 11–17). Available information, perceived family needs and life burdens, cultural patterns and traditions, emotional upsets, and physical reactions to dieting were reported in prior qualitative research (8–13, 16, 17) as well as in quantitative research (26), as were cognitive strategies of changing perspective (14), monitoring intake, and self-talk to encourage adherence or legitimize slips (15, 27). Self-efficacy predicted dietary change in one study (28), echoing the commitment, assertiveness, and willingness to try new things seen in adherent women studied here. Changes in cooking and eating behaviors linked by others (19, 29) to decreases in dietary fat consumption also were seen here and were reinforced by the instruction in the Women’s Health Initiative dietary modification intervention.
However, no previous investigators have detailed this range of distinctions in organismic influences between adherence and nonadherence. The a priori separation of environmental stimuli and internal attitudes and beliefs within the SORC framework required us to elicit and distinguish analytically social pressures that were overt and explicit from those that were merely assumed or expected. This approach enabled us to identify internalized social pressures, personal tastes, and habits as major barriers to adherence and the presence of external influences and supports as less important than the vulnerability of the individual.
Despite parallels with past research and the lack of age-related differences in this older sample, it is not yet known whether age is a determinant of the relative strength of particular influences on dietary change. It is notable that no women in this sample mentioned personal appearance as a motivator for reducing fat intake; older women may be less inclined to alter their diet for the sake of attractiveness than were younger samples (9, 12–15). The strength of personal food habits and customs may increase over time, particularly in aging women who have lost spouses and significant family members and may use food as a connection to a happier past. Finally, longevity itself may be a disincentive to change health habits (18); if one has survived to old age, there may be little perceived reason to undergo self-deprivation for the sake of a small qualitative improvement in health status.
The present design did not enable us to distinguish women whose commitment was strengthened by the WHI dietary modification intervention from those who came to the program with strong internal resources. These findings are further limited by the exploratory design, inability to assess interrater reliability in the entire data set, and lack of repeated contact with participants. As noted earlier, the possibility of misclassification of a few minority women with FFQ scores that approached the 25% cutoff point should be taken into consideration when judging the importance of these findings.
In future research, the external, organismic, and contingent factors identified here can be the basis for instrumentation to measure readiness or likelihood of successful dietary change. After appropriate testing, such a tool can be used to describe more precisely the influences on dietary behavior change in a larger sample of older individuals and explore differences between men and women and across the aging trajectory. It is not known whether long-standing habits or preferences are a greater burden for older women than for younger women or whether health concerns may be a stronger motivator for older women; a wider age range in sampling is called for to answer these questions. Future qualitative investigations should further explore the cognitive and emotional dynamics of making a long-term life change toward healthy eating and the kinds of support that enable older women to commit themselves to changing long-standing patterns of using high-fat foods in response to social pressure or emotions. Future research also is needed to sort out the processes involved in replacing the immediate reinforcing value of eating high-fat foods for cognitive reinforcers (eg, pride in accomplishing dietary change) and delayed gratification (eg, reduced health risk). In addition, it will be important to elucidate whether satisfaction with newly learned dietary choices precedes commitment to dietary change, whether initial commitment leads to higher satisfaction with dietary change, or whether both commitment to change and satisfaction with dietary choices must initially be present for dietary change to occur.
In the remaining years of the Women’s Health Initiative and in similar projects, the impact of tailoring support to organismic influences (ie, building commitment and skills to overcome challenges to dietary change) also merits qualitative and experimental investigation. In future interventions, the intensive phase of self-monitoring and frequent exposure to group support and palatable low-fat foods might be extended for a longer period or extended on a selective or contingent basis for women who have not yet made lasting life changes to support adherence.
In the meantime, clinicians can bear in mind that although the leading causes of death in postmenopausal women can be linked to poor dietary behavior (30), it is difficult for many older women to resist the social and internal expectation that they will provide comforting and indulgent food for their families and themselves. Health and nutrition professionals should continue to help women identify their emotional and social triggers for consuming high-fat foods and develop strategies to increase self-awareness and manage these situations. The next challenge will be to develop interventions that help women turn from high-fat foods as culturally endemic sources of nurturing and pleasure toward newly satisfying commitments to health over the longer term.
This publication was made possible by National Institutes of Health/National Institute on Aging Grant 5K01 AG00818. The contents of the article are solely the responsibility of the authors and do not necessarily represent the official views of the National Institute on Aging. The authors acknowledge the Women’s Health Initiative (WHI) centers (New York and Worcester) that participated in this study. We thank Sylvia Wassertheil-Smoller, PhD, Yasmin Mossavar-Rahmani, RD, PhD, Yvonne Raiford, and Marci Brown from the New York WHI center and Ingrid Lofgren, RD, Kathy O’Leary, and Nancy Polakowski from the Worcester center for their assistance in conducting this study.
1 When black and Hispanic women from the New York center were telephoned and invited to participate in focus groups, about half reported problems that would interfere with attendance, including conflicts with the scheduled time, lack of transportation, and responsibility of caring for a family member. Given the exploratory nature of the study, the desire to capture minority women’s viewpoints, and the relative adequacy of telephone interview data, the data collection protocol was expanded to include telephone interviews. Cited Here...
2 Although each data collection format had inherent limitations, adherent and nonadherent women were represented in the two formats in fairly equal proportions, and no differences were detected in the relative frequency of particular responses across the group and individual formats. Cited Here...
1. Mott JW, Wang J, Thornton JC, Allison DB, Heymsfield SB, Pierson RN Jr. Relation between body fat and age in 4 ethnic groups. Am J Clin Nutr 1999; 69: 1007–13.
2. Boyd N, Martin L, Beaton M, Cousins M, Krivkov V. Long-term effects of participation in a randomized trial of a low-fat, high carbohydrate diet. Cancer Epidemiol Biomarkers Prev 1996; 10: 117–38.
3. White E, Shattuck A, Kristal A, Uran N, Prentice R, Henderson M, Insull W, Moskowitz M, Goldman S, Woods M. Maintenance of a low-fat diet: follow-up of the women’s health trial. Cancer Epidemiol Biomarkers Prev 1992; 1: 315–23.
4. The Women’s Health Initiative Study Group. Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials 1998; 19: 61–109.
5. Kanfer F, Phillips J. Learning foundations of behavior therapy. New York: John Wiley & Sons 1970.
6. Chesney M, Feuerstein M. Behavioral medicine in the occupational setting. In: McNamara J, editor. Behavioral approaches to medicine. New York: Plenum; 1979. p. 267–90.
7. Keefe F. Assessment strategies in behavioral medicine. In: McNamara J, editor. Behavioral approaches to medicine. New York: Plenum; 1979. p. 101–29.
8. Currie C, Amos A, Hunt S. The dynamics and processes of behavioral change in five classes of health-related behavior: findings from qualitative research. Health Educ Res 1991; 6: 443–53.
9. McKie L, Wood R, Gregory S. Women defining health: food, diet and body image. Health Educ Res 1993; 8: 35–41.
10. Falk L, Bisogni C, Sobal J. Food choice processes of older adults: a qualitative investigation. J Nutr Educ 1996; 28: 257–65.
11. Nelson M. Health practices and role involvement among low-income working women. Health Care Women Int 1997; 18: 195–205.
12. Walcott-McQuigg J, Sullivan J, Dan A, Logan B. Psychosocial factors influencing weight control behavior of African American women. West J Nurs Res 1995; 17: 502–20.
13. Juarbe T. Cardiovascular disease-related diet and exercise experiences of immigrant Mexican women. West J Nurs Res 1998; 1998: 20: 765–82.
14. Johnson R. Restructuring: an emerging theory on the process of losing weight. J Adv Nurs 1990; 15: 1289–96.
15. Hamilton D. Continual monitoring: a theoretical model of the weight loss maintenance process [dissertation]. Berkeley (CA): University of California, Berkeley; 1988.
16. Lyons M. The phenomenon of compulsive overeating in a selected group of professional women. J Adv Nurs 1998; 27: 1158–64.
17. Popkess-Vawter S, Brandau C, Straub J. Triggers of overeating and related intervention strategies for women who weight cycle. Appl Nurs Res 1998; 11: 69–76.
18. Nigg CR, Burbank PM, Padula C, Dufresne R, Rossi J, Velicer WF, Laforge RG, Prochaska JO. Stages of change across ten health risk behaviors for older adults. Gerontologist 1999; 39: 473–82.
19. Keenan D, Achterberg C, Kris-Etherton P, Abusabha R, Von Eye A. Use of qualitative and quantitative methods to define behavioral fat-reduction strategies and their relationship to dietary fat reduction in the Patterns of Dietary Change Study. J Am Diet Assoc 1996; 96: 1245–50, 1253.
20. Tinker LF, Burrows ER, Henry H, Patterson RE, Rupp JW, Van Horn L. The Women’s Health Initiative. Overview of the nutrition components. In: Krummel DA, Kris-Etherton PM, editors. Nutrition in women’s health. Gaithersburg (MD): Aspen Publishers; 1996. p. 510–42.
21. Morgan D. Qualitative content analysis: a guide to paths not taken. Qual Health Res 1993; 3: 112–21.
22. Holsti O. Content analysis. In: Lindzey G, Aronson E, editors. The handbook of social psychology.Vol 1. 2nd ed. Reading (MA): Addison-Wesley; 1968. p. 596–692.
23. Downe-Wamboldt B. Content analysis: method, applications, and issues. Health Care Women Int 1992; 13: 313–21.
24. Knafl K, Webster D. Managing and analyzing qualitative data: a description of tasks, techniques, and materials. West J Nurs Res 1988; 10: 195–218.
25. Wilson H. Research in nursing. 2nd ed. Reading (MA): Addison-Wesley, 1989.
26. Palmeri D, Auld GW, Taylor T, Kendall P, Anderson J. Multiple perspectives on nutrition education needs of low-income Hispanics. J Community Health 1998; 23: 301–16.
27. Head S, Brookhart A. Lifestyle modification and relapse-prevention training during treatment for weight loss. Behav Ther 1997; 28: 307–21.
28. Shannon J, Kirkley B, Ammerman A, Keyserling T, Kelsey K, DeVellis R, Simpson RJ Jr. Self-efficacy as a predictor of dietary change in a low-socioeconomic status southern adult population. Health Educ Behav 1997; 24: 357–68.
29. Gorbach SL, Morrill-LaBrode A, Woods MN, Dwyer JT, Selles WD, Henderson M, Insull W, Goldman S, Thompson D, Clifford C. Changes in food patterns during low-fat dietary intervention in women. J Am Diet Assoc 1990; 90: 802–9.
30. Conn VS. Older women: social cognitive theory correlates of health behavior. Women Health 1997; 26: 71–85.