ABSTRACT

Objective: To explore how couples adjust to dietary man-agement of type 2 diabetes.

Design: Couples were interviewed, first together and thenseparately, during the first year after diagnosis and 1 year later.

Setting: Qualitative interviews conducted in hospital class-rooms using a semistructured interview guide.

Participants: Couples (N = 20) with a recently diagnosedspouse who met the study criteria were purposefullyselected from volunteers solicited from hospital-based dia-betes classes.

Phenomenon of Interest: Processes used by the couple toaddress the prescribed diet.

Analysis: Thematic analysis of interview transcripts usinggrounded theory to identify patterns of adaptation processesused over time.

Results: Three couple categories emerged (cohesive,enmeshed, and disengaged), representing adaptation to thediabetic diet. Initially, 5 couples were cohesive (teamworkapproach), 7 were enmeshed (nondiabetic spouse responsi-ble for the diet; spouse was dependent), and 8 were disen-gaged (spouses functionally separate; spouse was solelyresponsible for the diet management). A year later, themajority of couples were disengaged (n = 14), 1 coupleremained cohesive, and 4 couples remained enmeshed.Themes of flexibility, roles, rules, and communication var-ied across categories.

Conclusions and Implications: Understanding categories ofmarital adjustment to the diabetic diet may improve nutri-

tion-based diabetes interventions. Further study is needed toverify these findings in larger and more diverse populations.

KEY WORDS: diabetes, qualitative interviews, coupleinteractions, family systems theory

(J Nutr Educ Behav. 2005;37:226-234.)

INTRODUCTION

The prevalence of type 2 diabetes is approaching epidemicproportions in the United States.1 Adherence to recom-mended diets, the cornerstone of diabetes treatment,2,3 hasbeen identified as a major obstacle in diabetes management.4,5

When obstacles to dietary adherence are elicited, familyreactions (often termed family support) emerge as majorinfluences.6,7 Dietary treatment of type 2 diabetes disruptsfamily routines, especially mealtime routines,which are fairlyresistant to change.8 Some studies indicate that family sup-port is associated with better adherence to diabetic regimensand glycemic control9-12 and is the strongest, most consistentpredictor of adherence.13,14

Spousal interactions (ie, spousal support), a subset of fam-ily support, have a particularly strong impact on adjustmentto the diet because spouses are often intricately involved inmany of the tasks required for diet management.15,16 Spousalsupport has a major impact on food selection and mealplanning.16 Further, significant others’ perceptions of theimportance of following the regimen are a stronger predic-tor of the patients’ behavioral intention than are the patient’sown beliefs.17

However, no one definition of spousal support exists,18

especially one that includes negative spousal interactions.Although patient-perceived supportive behaviors fromspouses have been described, little is known of the maritalinteractions around adjustment to diet management.Theo-ries offer a framework for organizing and assessing what weobserve about behaviors.19 Family systems theory helpsexplain why family members behave as they do toward oneanother20 and is a logical choice to guide an investigation ofinteraction around diabetes diet management. The family

226

RESEARCH ARTICLE

Marital Interactions in the Process of Dietary Change for Type 2 Diabetes

DAISY MILLER, PHD, LN, CNS1; J. LYNNE BROWN, PHD, RD2

1Private practice, Riverdale, Maryland;2Department of Food Science,The Pennsylvania State University, University Park, Pennsylvania

This work was performed in partial fulfillment of requirements for a PhD thesisand was supported for 1 year by a US Department of Agriculture National NeedsFellowship.This work was conducted at the Deparment of Food Science,The Pennsylvania StateUniversity.Address for correspondence: J. Lynne Brown, PhD, RD, Department of Food Sci-ence,The Pennsylvania State University, 205A Borland Laboratory, University Park,PA 16802;Tel: (814) 863-3973; Fax: (814) 863-6132; E-mail: f9a@psu.edu.©2005 SOCIETY FOR NUTRITION EDUCATION

Journal of Nutrition Education and Behavior Volume 37 Number 5 September • October 2005 227

(or marital dyad) is not a group of individuals but a dynamicsystem in which each individual’s behavior affects every-body else. Additionally, family systems theory includesdimensions of cohesion (degree of emotional closeness),flexibility, and communication, all major coping strategiesfor stressors such as illness.21

Although few studies have used family-based theory tostudy adults with type 2 diabetes, a number have used fam-ily theory to examine family functioning and glycemic con-trol of type 1 diabetes in children and adolescents. Bothcross-sectional and longitudinal studies indicate that goodfamily functioning is associated with better adolescentglycemic control.11, 22-26

We chose family systems theory to guide our inquiry toshift our attention from the individual patient to thedynamic marital subsystem (the couple) in which diet man-agement occurs. Our subjects were couples in which onespouse had been recently diagnosed with type 2 diabetes.Weused qualitative methods to satisfy our research objectives: (1)to provide a rich description of the processes (mechanismsused to regulate member relationships such as role enact-ment and rules27) associated with dietary change over timefor diabetes at the couple level and (2) to group couples, ifpossible, according to observed marital adaptation patternsaround the diabetic diet.

METHODS

Study Design

We used a longitudinal design of 2 sequential interviews,1 year apart, with 20 couples.The initial interview occurredwithin 1 year after diagnosis.The second interview occurreda minimum of 1 year following the first.The purpose of theinitial interview was to outline the framework of adjust-ment, whereas that of the second interview was (1) to builda richer description of the themes centered on the couples’adaptation to the diabetic diet and (2) to examine reportedchanges that occurred in the couples’ behavior over time toyield process-oriented themes.

Participant Recruitment

Volunteers were recruited through multiple visits to 4 out-patient hospital diabetes education programs meeting thestandards of the Pennsylvania Department of Health Dia-betes Control Program and serving predominantly Cau-casians of European descent in central Pennsylvania. Threehospitals were the major diabetes care centers for 2 countieswith metropolitan areas and the fourth served 1 rural county.All 3 counties were contiguous.We explained the study, dis-tributed informational flyers, and circulated a sign-up sheetat the beginning of class. All volunteers were contacted bytelephone to provide more study information and to deter-mine eligibility. Participation criteria were that (1) diabeteswas diagnosed within 1 year of recruitment; (2) the person

with diabetes was 45 years or older, based on increasedprevalence above this age28; (3) the couple was married orcohabiting; (4) both partners agreed to be interviewed; and(5) at least the patient had completed the outpatient sessionson dietary management. It took 10 months to recruit 20couples from 24 meetings (on average, 14 participants werepresent) using purposeful sampling to secure equal numbersof men and women with diabetes and a mix reporting suc-cess or difficulty with the diet. Many attending classes didnot meet eligibility criteria (ie, they were single, divorced,widowed, or more than 1 year past diagnosis).The cooper-ating diabetes educators estimated that half of the class par-ticipants were married and only 40% of those were newlydiagnosed. Of those estimated as eligible (N = 67), 60%(n = 40) met our criteria. Half of these met purposeful cri-teria for the study. The Office for Research Protections atThe Pennsylvania State University and the participating hos-pitals’ research review committees approved all procedures.

Interview Protocol

Interviews were conducted in participating hospital meetingrooms, except one done in the couple’s home at theirrequest. The couple interview protocol, including taperecording, was explained by telephone, and the subsequentinterview was scheduled at a time convenient for the partic-ipants. At the initial interview, all participants signed aninformed consent form granting permission to tape recordthe conversation and then completed a demographic ques-tionnaire that recorded age, height, weight, educational level,income range, and ethnicity. One interviewer, trained duringa pilot study, conducted all of the interviews using a semi-structured interview guide. First, the interview with bothspouses took place, typically lasting an hour.Questions to thecouple included open-ended queries about meals before thediabetes diagnosis (eg, How did the cooking work before thediagnosis?) and after the diagnosis (eg, How do meals worknow? Describe a typical day.). Blood glucose monitoring wasnot directly assessed. Separate interviews with each spousefollowed and lasted about 30 minutes, with the order deter-mined by a coin toss. Questions to the spouses focused onbehaviors that contributed in some way to diet manage-ment. For example, those spouses with diabetes were asked,“What kinds of things does your spouse do or say that helpyou follow your meal plan?” Likewise, spouses without dia-betes were asked,“What kinds of things do you do or say tohelp your spouse follow his/her meal plan?”At the conclu-sion of the 3 interviews, the couple received a cash gift.Theinterviewer requested and received permission to contact allof the couples 1 year later regarding the second interviews.

The second interview guide included both general andspecific follow-up of initial findings.The couples were con-tacted again by telephone, about 1 month prior to the antic-ipated date of the second interview. All but one couple(n = 19) agreed to participate. Reasons for nonparticipationwere not disclosed. The initial interviews were conducted

228 Miller and Brown/MARITAL INTERACTIONS IN PROCESS OF DIETARY CHANGE FOR DIABETES

from June 2000 to March 2001 and the second interviewsfrom July 2001 to March 2002. The supervising researchermonitored interview quality through review of typed tran-scripts as interviewing progressed. Data tapes were tran-scribed verbatim.

Data Analysis

Two researchers analyzed these transcribed data indepen-dently using 2 different styles of constant comparison.29 Eachindependently examined the initial interview data followedby comparison of findings and discussion, repeating thisprocess for the second interview data. Both researchers firstcoded themes (eg, food choice alterations), which were laterclustered and developed into suprathemes (eg, flexibility)and ultimately couple categories that described patterns ofcouple interaction around the prescribed diets. However, 1researcher (who also conducted the interviews) developedthemes using intensive note taking while conducting pur-poseful sampling (ie, seeking couple situations to fill gaps inunderstanding). The second researcher extracted data thataddressed broad coding themes from each finished interviewand compared these with all other interviews at that timepoint. Based on the thematic patterns, each researcher inde-pendently placed couples into 1 of 3 categories. Becausecomparison revealed 80% agreement in categorization forinitial interview data but only 65% to 70% agreement for thesecond interview data, all data were systematically reassessedindependently using mutually agreed on checklists of themesand subthemes of dietary interaction and constant compar-ison within initial and then second interview data sets.Com-parison of categories and themes ultimately reached 90% to95% agreement for both time points, and subsequent discus-sion resulted in mutual agreement in final couple catego-rization.The degree of successful dietary management wasbased on reported food choices, portion control, and mealplanning, food shopping, and food preparation patterns(unpublished data).

RESULTS

Ten couples represented each sex of diabetic spouse at the ini-tial interview; 1 couple with a diabetic female declined toparticipate at the second interview. Participants were Cau-casian and had a mean age of 65.6 ± 6.9 years for diabeticmales and 55.9 ± 7.1 years for diabetic females. Couples withdiabetic men were married for an average of 31 ± 18 years,and couples with diabetic women (n = 9) were married anaverage of 30 ± 10 years (1 couple had cohabited for 18 years).Couples’ reported combined incomes were < $20 000 (n = 7),$20 000 to $40 000 (n = 7), and > $40 000 (n = 5). Themajority of spouses (78%) had a high school education or less.

Although all diabetic spouses faced similar challengesadjusting to the prescribed diet (eg, learning the exchangesystem), couples used different processes to address these

challenges. Data analysis revealed that couples fell into 1 of3 core categories (cohesive, enmeshed, or disengaged) basedon the couples’ approach to the prescribed diet. Four majorthemes emerged within the couple categories: flexibility (theextent to which spouses made diet changes), roles (assigningand implementing jobs for diet management, such as mealplanning), rules (setting guidelines for conduct around thediet), and communication (open or closed verbal exchangebetween spouses about the diet). In open communication,discussions were free flowing and 2 way, with little conflictbecause disagreements were respected and resolved, whereasin closed communication, discussion was either absent or 1way, often with unresolved conflict.19 The couple’s commu-nication skills and the degree to which they exhibited flex-ibility, altered roles, and assigned rules were the basis of theirassignment to a core category of adaptation.

Observations at Initial Interviews (N = 20 Couples)

Cohesive couples (n = 5 couples; 3 females and 2 maleswith diabetes).

I think people have to realize that you got to work together onsomething like this. It becomes a…it becomes a full-timejob,…and I think that people have to realize that everyone inthe family has to work for the person that has it.

In all cohesive couples, both spouses worked together asa team to make dietary changes, although there was somevariability in the extent of teamwork. Diabetic spouses tookresponsibility for their food choices, and nondiabetic part-ners were actively engaged in establishing and sharing theprescribed eating pattern.

Flexibility. In cohesive couples, both spouses were flexibleand made extensive dietary changes for meals and snacks.One diabetic spouse described her husband’s flexibility:“Every Thursday it’s payday, so he brings me strawberriesand grapes or bananas…where before he’d bring home chipsand popcorn and that. He brings me home the things I’msupposed to have now.” In implementing the diet regimen,both spouses became involved in meal planning, countedexchanges, changed mealtimes, chose low-fat foods, stoppedentry of treat foods into the home, altered cooking methods,changed restaurant patronage, read labels, cut down portionsizes, followed new recipes, and changed meal serving styles(eg, served food portioned on plates instead of using servingbowls). Many of these diabetic spouses felt that successfuldietary change occurred because their spouses had made sig-nificant changes in their own diets.

Roles. Traditional roles (eg, meal planning, shopping, andcooking) were negotiated and subsequently altered postdi-agnosis to address the new challenges of diabetes manage-ment. Cohesive couples shared more of these diet manage-ment roles (eg, both spouses planned meals) than enmeshedand disengaged couples. Meal planning, a critical role,

Journal of Nutrition Education and Behavior Volume 37 Number 5 September • October 2005 229

appeared to require the most teamwork because the abilityto understand the meal plan required, for some, each other’shelp. Successful meal planning involved working together todecipher, understand, and make cooperative decisions aboutfood choices.Although sharing roles was not absolutely crit-ical in the process of creating teamwork, it did help manycouples implement the diet regimen.Only 2 diabetic spousesmentioned efforts at routine blood glucose testing, indicat-ing that this was their responsibility.

Rules. Negotiation of rules, facilitated by clear communica-tion, was important in promoting teamwork and changingshared diet patterns. Setting rules most often involved nego-tiating what treat foods entered and were eaten in the house,what food was served during family (especially grandchil-dren’s) visits, meal scheduling, meal preparation methods, andrestaurant selection. In most cohesive couples, rule settingwas continuously practiced, and subsequent rules were gen-erally clear and well-defined. For example, one couple triedkeeping potentially tempting food items in the house undera “trial basis.” If the diabetic spouse found the food item tootempting, both spouses agreed that the food would no longerbe brought home. Setting rules as opposed to relying onassumptions was important because it was not always obviouswhich foods were considered tempting.No single set of ruleswas appropriate for all cohesive couples; each couple neededto set rules suited to each spouse’s unique preferences.

Communication. In general, communication in cohesivecouples was open. Spouses freely discussed eating behaviorsand food choices, and the diabetic spouse accepted dietarymonitoring comments without feeling resentful. Monitoringcomments about food selection or quantity taken were oftenconsidered “suggestions” or “gentle reminders,” and howsomething was said was more important than what was said.Monitoring preferences varied across couples. For example,saying, “You shouldn’t be eating that” was considered puni-tive by some spouses but was welcomed by others.To avoidresentment of reminders, some diabetic spouses would granttheir partners permission to monitor. One diabetic spousereported coming home from work and saying,“I had a stress-ful day today;watch me tonight… If you think I’m doing toomuch, I want you to let me know.” In general, talking aboutmundane issues (eg, mealtimes) and more difficult topics (eg,monitoring) facilitated flexibility, shared roles, and clear rules.

Enmeshed couples (n = 7 couples; 2 females and 5 males with diabetes).

It’s been hard on me [nondiabetic]…mainly controlling whathe eats falls pretty much on me…

In enmeshed couples, nondiabetic spouses were responsi-ble for managing the diet and diabetic spouses were depen-dent on their efforts. Some diabetic spouses acted helplessand needy, reflecting a lack of initiative or interest in theirown diet management. Some nondiabetic spouses who

appeared to desire parental-like control seemed to encour-age this dependence.

Flexibility. Spouses without diabetes generally initiated andcarried out the required dietary changes,motivated sometimesby improving their own health. One explained: “So I don’tkeep a lot of stuff in the refrigerator anymore.Now I make Jell-O and low-fat puddings and stuff like that. I more or less fol-low his diet too.We ate a lot of sweet stuff. I baked cakes andthen we had a lot of cookies and pies; I don’t do that anymore.”

However, some refused to give up preferred but temptingsnacks, often eating these when their partner was not around(eg,“I don’t have [diabetes]. I realize I shouldn’t be having apiece of chocolate or having a cookie occasionally, but whenI do, I try not to eat it in front of him.”) Spouses with dia-betes typically resisted change and were unwilling to acceptall of their partner’s interpretations of the diet rules. Onenoted:“She’ll say,‘Well, you’re not suppose to do this,’ and I’llsay,‘Hey, wait a minute; that wasn’t said at the [diabetes class]meeting.” Some ate what was presented at meals yet overateportions, refusing to accept total spousal control of their foodintake.One said:“I get angry. I don’t really say anything; I justget angry. He just knows I’m angry because I can’t eat some-thing. I’ll say,‘Why can’t I eat this?’ He’ll say,‘Well, [because]you had this and you had that.” Some also rebelled by eatingtreat foods in the house when their spouses were not look-ing. One man explained that his wife counts his potato chipsfor lunch and said,“Now, if there’s potato chips in the drawer,I do cheat on the potato chips, like when I get up and I knowthe potato chips are there, I’ll grab a few.” In general, nondi-abetic spouses were more flexible about dietary change thandiabetic spouses, who often resented and disagreed with thedietary rules their spouses imposed.Two diabetic spouses usedacceptable blood glucose values to resist change in foodchoices, whereas several others resisted their spouses’ effortsto instill consistent blood glucose monitoring.

Roles. Roles in enmeshed couples were not shared, and lit-tle alteration of roles (often traditional) was evident. In 5 ofthe 7 couples, the nondiabetic wife managed the meal plan-ning, shopping, and cooking, whereas in the other 2 couples,the nondiabetic husband did this. Because nondiabeticspouses had to balance the diabetic’s taste preferences (manywere picky eaters) with the diet guidelines with little or nohelp from their partner, meal planning was particularly chal-lenging. Some diabetic spouses appeared to be pleased withrole assignments,whereas many nondiabetic spouses felt bur-dened by their responsibility as diet managers. Imbalance ofand dissatisfaction with role assignments may have discour-aged some nondiabetic spouses from planning meals in strictaccordance with the diet guidelines. In 5 couples, the non-diabetic partner set out or cared for the glucometer andmonitored measurements and results.

Rules. Enmeshed spouses often disagreed on the rules ofdietary conduct because many diabetic spouses thought that

230 Miller and Brown/MARITAL INTERACTIONS IN PROCESS OF DIETARY CHANGE FOR DIABETES

the nondiabetic’s rules were too rigid. For example, one non-diabetic wife reported that her husband was allowed exactly1 cup of pasta, whereas her husband thought it was acceptableto “eyeball”portions.Spouses without diabetes often attemptedto enforce rules about eating in social situations (eg, restaurantsor family gatherings), and spouses with diabetes rebelled byovereating and making poor food choices.Conflict made rulesetting inconsistent, and any rules were short-lived.To mini-mize conflict, some nondiabetic spouses attempted to set ruleswithin the realm of what was acceptable to their spouse, soft-ening and limiting compliance.Thus, rule setting in enmeshedcouples was ineffective in facilitating dietary change.

Communication. In general, communication in enmeshedcouples was closed because spouses did not freely discuss thediabetic diet or effectively negotiate the delivery mode forreminders and suggestions. Dietary monitoring commentswere often resented and viewed as “nagging,” punitive, orinstilling guilt. Some diabetic males felt that their wives weretoo “motherly” when they explained the meal plan rules.Spouses with diabetes strongly discouraged communicationin public situations because many believed that public mon-itoring of food choices was embarrassing. Closed communi-cation appeared to contribute to the maintenance of imbal-anced roles and hindered the process of rule negotiation.

Disengaged couples (n = 8 couples; 5 females and 3 males with diabetes).

I guess I just get used to doing it on my own instead of depend-ing on somebody for support…

In disengaged couples, the diabetic spouse was solelyresponsible for his or her diet management. At the initialinterview, only 2 highly independent, self-motivated dia-betic wives were making progress in dietary change. In gen-eral, nondiabetic spouses either acted passively uninterestedin (ie, ate whatever partner prepared with little expressedinterest in changes) or sabotaged diet management efforts(eg, refused to alter food choices or continued to bake won-derful desserts).

Flexibility. We observed several patterns of flexibility in dis-engaged couples.These were (1) both spouses were inflexi-ble and neither made dietary changes; (2) the diabetic spousewas flexible, independently making dietary changes, and thenondiabetic spouse was not; (3) the diabetic spouse was flex-ible and the nondiabetic spouse was only partially flexible(eg, cooperating by eating whatever was prepared at mealsbut then overindulging in favorite snacks); and (4) the dia-betic spouse was partially flexible, carefully following therecommended meal plan for only part of the day (usuallybreakfast and lunch), and the nondiabetic spouse was inflex-ible. Common changes for flexible diabetic spouses werealtered meal planning (mainly using the exchange system)and more self-monitoring of food choices and portion sizes(both at home and in social situations). Restaurant eating,

social visits, and mealtimes were often scheduled aroundnondiabetic spouses’ preferences. Other less motivated dia-betic spouses made very few or no dietary changes.Althoughflexibility varied considerably within the disengaged group,the responsibility for initiating and implementing changesalways rested on the diabetic spouse, whose decisions wereinfluenced by their spouse’s flexibility.

Roles. These were generally not shared in disengaged cou-ples, although several fairly independent diabetic men tookon new roles postdiagnosis (eg, shopping or planning meals)so that they could have more control of their diets. How-ever, traditional roles did not change in couples with lessmotivated diabetic spouses. Although meal planning wasgenerally considered difficult, some of those who had dia-betes and were independent discouraged any involvementof their spouses. Others, finding meal planning overwhelm-ing, wanted more assistance from their spouses and feltresentful when little help was offered and the spouse’s sab-otaging behavior caused conflict. Disengaged couples’inability to develop shared roles may have discouraged somediabetic spouses from addressing the meal plan challenges.Four diabetic spouses indicated that they monitored theirblood glucose, although the frequency varied.

Rules. In disengaged couples, rule setting was either incon-sistent or nonexistent.When spouses did not set or negoti-ate rules, meal planning was minimal and each spouse did hisor her “own thing” for meals. Mealtimes were often random,food shopping involved more impulse buying than planning,and restaurant choices were based on taste and conveniencerather than diabetic needs. Inappropriate treat foods werestored and eaten in the home, often generating conflict, butrules to govern this were inconsistent or absent.One diabetichusband reported: “…She’s probably the worst thing thatcould happen to my diet…; she loads the house up withcandy…. If there’s no candy around, I wouldn’t eat it.” Somecouples seemed to have good intentions but lacked the com-munication skills needed to negotiate rules effectively.Whenrules around the diet were not consistent, it was difficult forthe diabetic spouse to implement dietary changes.

Communication. Disengaged couples had a difficult timediscussing meal plan management, and communication waspredominantly closed. Blaming and accusations were com-mon because each spouse lobbied for his or her own inter-ests, for example, preferred foods, meals, and mealtimes.When nondiabetic wives with a history of weight problemsfelt that their diabetic spouses had been insensitive to theseproblems, conflict, jealousy, and uncooperativeness oftenemerged around the diet. Dietary monitoring comments,although typically infrequent, were considered disrespectfuland punitive. Sometimes diabetic spouses rebelled againstmonitoring comments by overeating. Closed communica-tion in disengaged couples hindered their ability to negoti-ate roles and develop consistent diet management rules.

Journal of Nutrition Education and Behavior Volume 37 Number 5 September • October 2005 231

Observations at the Second Interviews (n = 19 Couples)

Initial cohesive couples (n = 5). Continuing cohesive cou-ples (n = 1 couple, 1 female with diabetes). At the secondinterview, 1 couple remained cohesive, despite some shifttoward disengagement.They remained cohesive because thenondiabetic husband was partially flexible and had made somelifestyle changes (instigated by the spouse with diabetes), thediabetic wife was responsible for most of the diet managementroles (although meal planning was still partially shared) andmeasuring blood glucose, rules were established that were stillacceptable to both, and the couple still had open communi-cation. Increased disengagement was due to the diabetic wife’staking more control of diet management while also revertingto stress-driven poor eating habits.The cohesive pattern sur-vived at the second interview with somewhat less dietarycompliance because most changes implemented by the initialinterview were acceptable to both partners.

Shifts: cohesive to disengaged (n = 4 couples, 2 females and2 males with diabetes). Four of the 5 couples who were cohe-sive at the initial interview shifted to disengaged at the second.These spouses became less flexible, resorting to old eating pat-terns and paying less attention to meal plan guidelines. Forsome, family or financial problems took precedence over dia-betes management. Both spouses’ roles changed as well. Somespouses fell back into traditional roles that had been discon-tinued soon after the diagnosis (initial interview), such as thatof baker. Other spouses shifted from active (initial interview)to limited involvement in meal planning (second interview).All diabetic spouses were measuring blood glucose, but withwidely varying frequency. Many of the rules that were estab-lished by the initial interview fell apart by the second becausefew rules were revised and few new rules were set. Rules dis-integrated (especially about handling treat foods), most oftenwhen life was most stressful (eg, during holidays).When cou-ples lacked rules, meals, snacking, eating out, and meal sched-uling were decided impulsively and/or by convenience.Com-munication diminished, becoming closed, as some diabeticspouses reported that discussing diet management was nolonger necessary, despite regression to old eating habits.

Despite these changes, 2 diabetic men maintained rea-sonably compliant diets because they had “tremendous”willpower and couple adjustment did not require majorchanges in roles, rules, and food choices. One couple hadmade changes in food choices 5 years previously owing tothe diabetic husband’s heart attack, and he performed all ofthe food chores at initial and second interviews. In the other,the wife did not alter her cooking pattern because their cus-tomary diet was fairly balanced at the initial interview andthe diabetic husband knew his exchange pattern and had thewillpower to resist significant in-home temptations and toexercise strong portion control at the second interview.However, newer rules did deteriorate and communicationdiminished in both couples.

Initial enmeshed couples (n = 7). Continuing enmeshedcouples (n = 4 couples, 1 female and 3 males with diabetes).Although diabetic spouses were still dependent on theirpartners for meal planning, we noted a shift to disengage-ment as enmeshed spouses became less flexible, paying lessattention to planning meals according to the diet guidelinesand to controlling portions and food choices (especially treatfoods). Responsibility for the roles of meal planning, shop-ping, and cooking did not change, but diligence in executiondropped, and most nondiabetic spouses reported feeling bur-dened and frustrated. Nondiabetic partners remainedinvolved in glucose monitoring, which some spouses stilldisregarded. Disagreements about meal plan rules were morecommon at the second interview. Many diabetic spouses dis-missed their partner’s rules, feeling that they were being“babied.” Communication became more limited (closed),and some nondiabetic spouses stopped dietary monitoringowing to partner resentment. Compliance dropped, but theenmeshed pattern survived owing to diabetic dependenceand nondiabetic spouses’ willingness to continue a lesser pat-tern of support. Only 1 couple with a highly committednondiabetic wife and very compliant diabetic husband main-tained acceptable dietary patterns in the home that dimin-ished in other social situations. The husband’s earlier heartattack had resulted in some dietary changes prior to his dia-betes diagnosis.

Shifts: enmeshed to disengaged (n = 3; 1 female and 2 maleswith diabetes). Three of the 7 couples who were enmeshedat the initial interview shifted to disengaged at the second.Both spouses became less flexible, reverting to traditionaleating patterns. In several, we observed an almost completereversion to old meal choice patterns and eating habits. Rolerelinquishment and claiming independence were majorthemes. Nondiabetic spouses returned to prediagnosis roles,which often included baking and cooking large, rich, tradi-tional dinners. Diabetic spouses took control of blood glu-cose readings but used the results to justify poor choices orto reduce dietary efforts. Disagreements about the meal planrules and blaming became more frequent and intense as dia-betic spouses sought more independence. Communicationbecame more closed as problems such as emotional eatingand purchasing and baking tempting foods sparked conflict.Some couples reported feeling guilty for abandoning thediet, whereas others were defensive, especially nondiabeticspouses who relinquished their roles as the meal plan man-agers, blaming the diabetic spouses for reverting to old ways.Dietary compliance was poor for these couples.

Initial disengaged couples (n = 7 of 8). Continuing dis-engaged couples (n = 7 couples;4 females and 3 males with dia-betes). All couples assessed as disengaged at the initial inter-view were also assessed as disengaged at the second. Mostwho had diabetes and remained disengaged paid less carefulattention to their diets as the year progressed (continuedinflexibility). However, a few who had diabetes were more

232 Miller and Brown/MARITAL INTERACTIONS IN PROCESS OF DIETARY CHANGE FOR DIABETES

independent and made some effort to follow their dietsdespite marital barriers. Role delineation did not change.Five diabetic spouses reported using blood glucose monitor-ing to varying degrees.Disagreements around the rules of themeal plans, however, intensified. Communication was char-acterized by marital tension and conflict, especially with thediagnoses of other diseases and during periods of high stress.Only 3 highly motivated and skilled diabetic wives demon-strated some dietary success because they did all of the foodchores, had a partner who would eat anything without com-plaint, or catered to their husband’s preferences by makingseparate meals or dishes for themselves. Two used multipledaily blood glucose readings to control highs and lows.Theydid not challenge flexibility, negotiate new rules (for snacksor treats), or alter roles, so communication was not required.By shouldering the entire burden of diet management, thesewomen caused minimal change in couple interactions.

DISCUSSION

This study provides an in-depth examination of dynamicmarital interaction patterns emerging around dietary man-agement within 2 years of diagnosis, a period when an inter-est in altering dietary patterns may be highest. Using familysystems theory as the framework, our analysis revealed 3 cat-egories of marital adaptation, cohesive, enmeshed, and dis-engaged, a pattern observed in families coping with otherchronic conditions.30 Couples in each category faced thesame challenges but handled these differently. Our findingshighlight the influence of both positive and negative inter-actions on the process of dietary change31 and indicate howimportant spousal flexibility and communication skills arefor successful adaptation to the diabetic diet. However, thesustainability and degree of successful dietary complianceemerging within each category varied.

Only one couple sustained the cohesive pattern despitebecoming more disengaged. Immediately after diagnosis, in“the honeymoon period,” the original cohesive coupleswere highly motivated to work together, but this survivedonly if both partners were flexible and changes in roles andrules were acceptable to both. Still this initial “teamwork”appeared to foster emergent success in disengagement at thesecond interview for a few previously cohesive diabetic men.

Enmeshment was moderately sustainable and, based onthe medical history of the one enmeshed couple reportingsome success in diet management at the second interview,may reflect established roles that are very resistant tochange.32 Olson reported that enmeshed families could func-tion well as long as all family members accept the rules androles.33 Spouses who were enmeshed at the initial interviewand moved to disengaged at the second were unhappy withtheir roles and the imposed rules. Poor negotiation skills(causing disagreement about rules) appeared to facilitate thisshift. Imposition of rules and excessive monitoring can back-fire unless the diabetic spouse is comfortable with being both

dependent and acquiescent. Moving from enmeshment todisengagement did not result in successful dietary manage-ment in our sample.

Disengagement was sustained by poor communicationand limited ability to negotiate, set and follow rules, or alterroles that addressed changing circumstances (eg, employ-ment status, family problems, holidays). It became the pre-dominant adaptation pattern at the second interview. Thiswas also the predominant pattern in a cross-sectional assess-ment of 247 couples with a diabetic spouse who volunteeredto complete several instruments measuring general familyfunctioning.34 Diabetic spouses in 5 disengaged couplesexhibited some success in dietary management (2 previouslycohesive and 3 continuing disengaged). These diabeticspouses were independent, believed that the disease wassolely their responsibility, and had great willpower to resisttemptation. Their partners were passively compliant withalterations in food choices or the diabetic spouse did notchallenge their preferences. No changes in rules or roleswere implemented because the diabetic spouse shoulderedall responsibility for change. The majority (4/5) controlledthe family food system via their responsibility for foodchores. Thus, the couple interaction pattern was not chal-lenged or changed. However, at the second interview, eventhese diabetic spouses faced difficulties (eg, less cooperativespouses, family problems, increased stress), which did notbode well for future compliance.

Adaptation to dietary change in the past owing to heartdisease did facilitate some subsequent successful adaptationto diabetes, but not in all cases. By the second interview, allbut 3 couples indicated that blood glucose monitoring wasinvolved in diet management but use ranged from severaltimes a day to several times a month. Only 3 diabetic spouseswere using this to regulate daily highs and lows (1 cohesiveand 2 “successful” disengaged).

Within systems theory, the master trait that determineshow families cope or function in the face of stressful eventsis adaptability.35,36 In our sample, the 3 patterns emerging by2 years postdiagnosis of diabetes indicated that couple adapt-ability (as expressed through the degree of flexibility in foodchoices, rules, and roles) plus communication skills deter-mined the couple adjustment pattern. Enmeshed and disen-gaged patterns appeared to represent 2 opposite extremes offirst-order change, that is, patterns that retain the original sys-tem characteristics with minimal alterations in couple inter-actions.The cohesive pattern, similar to the teamwork patternoutlined by Lyons et al,30 might lead to second-order changeand eventual morphogenesis37 to a new pattern of interactionaround food, that is, a change in the family system itself.

Families develop patterns of interaction over time (calledmaturation) that reflect degrees of individualization38 based onboundaries (differentiating the individual from others) gov-erned by relational rules.The degree of individuation was lowin the enmeshed and high in the disengaged patterns. Theeventual mature pattern of couple adjustment to type 2 dia-betes may require more than 2 years to develop. However, the

Journal of Nutrition Education and Behavior Volume 37 Number 5 September • October 2005 233

prevalence of the 3 patterns observed will undoubtedly beaffected by access to or training in process skills (mechanismsused by families to change, regulate,or maintain member rela-tionships) that encourage a level of teamwork satisfactory forboth spouses.We hypothesize that our cohesive pattern couldbe a prerequisite for the lifestyle change needed to improveoutcomes for diabetic patients living with a spouse and that itwill be less prevalent than the other 2 patterns unless thesecouples receive more skills training and professional supportthan the current system appears to provide.

IMPLICATIONS FOR RESEARCH AND PRACTICE

Study limitations include a small, nationally unrepresenta-tive sample with limited socioeconomic and no racialdiversity, some variation in time since diagnosis, possiblebias in assessing couple adaptation categories, and noexamination of the nutrition and medical managementinterface. Exposure to intensive medical nutrition therapymight increase compliance.39

However, our research provides a basis for further workdesigned to understand couple adaptation to diets for type2 diabetes. Based on our findings, we suggest the following:

• The adjustment patterns observed must be verified byexperimental studies using larger, more representativesamples. Valid scales that allow identification of adjust-ment patterns must be developed to quantify their preva-lence. Such studies could examine how the interface ofmedical and nutrition management and intensive medicalnutrition therapy affect these patterns.

• Many couples reverted to or relied on old interpersonalpatterns and communication skills to cope with dietarychange. Interventions designed to build process skills(negotiation of rules and monitoring patterns, resolvingconflicts, sharing responsibility, and encouraging flexi-bility and trade-offs) among couples in which one part-ner has type 2 diabetes need to be designed and tested.The intervention objectives might be the prevention ofextremes of adjustment arising from a lack of rules, bur-densome roles, unsatisfactory monitoring systems, andinflexibility.

• Where possible, diabetes educators should consider work-ing with both a newly diagnosed diabetic patient and hisor her spouse. Although educators encourage spouses ofparticipants in hospital diabetes classes to attend theseclasses, these educational programs do not actively incor-porate nondiabetic spouses into course activities. How-ever, including both spouses in individual dietary coun-seling sessions may offer opportunities to suggest sharingroles, building flexibility, establishing rules, and improvingcommunication patterns that will inspire further cooper-ation by spouses.

ACKNOWLEDGMENTS

This work was performed in partial fulfillment of require-ments for a PhD thesis and was supported for 1 year by a USDepartment of Agriculture National Needs Fellowship.

REFERENCES

1. Mokdad AH, Bowman BA, Ford ES,Vinicor F, Marks JS, Koplan JP.The

continuing epidemics of obesity and diabetes in the United States.

JAMA. 2001;286:1195-1200.

2. Boehm S, Schlenk E, Funnell M, Powers H, Ronis D. Predictors of

adherence to nutrition recommendations in people with non-insulin-

dependent diabetes mellitus. Diabetes Educ. 1997;23:157-165.

3. Garber A.The end of “easy.” Clin Diabetes. 1997;15:193-196.

4. Lockwood D, Frey M, Gladish N, Hiss R.The biggest problem in dia-

betes. Diabetes Educ. 1986;12:30-33.

5. Woodcock A, Kinmonth A. Patient concerns in their first year with type

2 diabetes: patient and practice nurse views. Patient Educ Couns.

2001;42:257-270.

6. Glasgow R. Social-environmental factors in diabetes: barriers to dia-

betes self-care. In: Bradley C, ed. Handbook of Psychology and Diabetes.

New York, NY: Routledge; 1994:335-349.

7. Cárdenas L,Vallbona C, Baker S,Yusim S.Adult onset diabetes mellitus:

glycemic control and family function. Am J Med Sci. 1987;293:28-33.

8. Mengel M.Functional assessment of families with a diabetic person.Pri-

mary Care. 1988;15:297-310.

9. Gerstle J,Varenne H, Contento I. Post-diagnosis family adaptation influ-

ences glycemic control in women with type 2 diabetes mellitus. J Am

Diet Assoc. 2001;101:918-922.

10. Anderson B. Diabetes and adaptation in family systems. In: Holmes C,

ed. Neuropsychological and Behavioral Aspects of Diabetes. New York, NY:

Springer-Verlag; 1990.

11. Hauser S, Jacobson A, Lavori P, et al. Adherence among children and

adolescents with insulin-dependent diabetes mellitus over a four-year

longitudinal follow-up: immediate and long-term linkages with the

family milieu. J Pediatr Psychol. 1990;15:527-542.

12. Hanson C, Henggeler S. Model of associations between psychosocial

variables and health-outcome measures of adolescents with IDDM.

Diabetes Care. 1987;10:752-758.

13. Glasgow R, Toobert D. Social environment and regimen adherence

among type II diabetic patients. Diabetes Care. 1988;11:377-386.

14. Wilson W, Ary D, Biglan A, Glasgow R,Toobert M, Campbell D. Psy-

chosocial predictors of self-care behaviors (compliance) and glycemic

control in non-insulin-dependent diabetes mellitus. Diabetes Care.

1986;9:614-622.

15. Edelstein J, Linn M.The influence of the family on control of diabetes.

Soc Sci Med. 1985;21:541-544.

16. Savoca M, Miller C. Food selection and eating patterns: themes found

among people with type 2 diabetes mellitus. J Nutr Educ. 2001;33:

224-233.

17. Shenkel R, Rogers J, Perfetto G, Levin R. Importance of ‘significant

others’ in predicting cooperation with diabetic regimen. Int J Psychiatry

Med. 1985;15:149-155.

234 Miller and Brown/MARITAL INTERACTIONS IN PROCESS OF DIETARY CHANGE FOR DIABETES

18. Hupcey J. Social support: assessing conceptual coherence. Qual Health

Res. 1998;8:304-318.

19. Noller P, Fitzpatrick M. Communication in Family Relationships. Engle-

wood Cliffs, NJ: Prentice-Hall; 1993.

20. Fingerman K, Bermann E.Applications of family systems theory to the

study of adulthood. Int J Aging Hum Dev. 2000;51:5-29.

21. Patterson J. Integrating family resilience and family stress theory. J Mar-

riage Fam. 2002;64:349-360.

22. Minuchin S, Rosman G, Baker L. Psychosomatic Families. Cambridge,

Mass: Harvard University Press; 1978.

23. Auslander W,Anderson B, Bubb J, Jung K. Risk factors to health in dia-

betic children: a prospective study from diagnosis. Health Soc Work.

1990;15:133-142.

24. Klemp S, LaGreca AJ.Adolescent with IDDM: the role of family cohe-

sion and conflict [abstract]. Diabetes. 1987;36(Suppl 1):18A.

25. Anderson B, Miller J,Auslander WF, Santiago J. Family characteristics of

diabetic adolescents: relationship to metabolic control. Diabetes Care.

1981;4:586-594.

26. White K, Kolman M,Wexler P. Unstable diabetes and unstable families:

a psychosocial evaluation of diabetic children with recurrent ketoaci-

dosis. Pediatrics. 1984;73:749-755.

27. Broderick CB. Understanding Family Process: Basics of Family Systems The-

ory. Newbury Park, Calif: Sage Publications; 1993.

28. Kenny SJ, Aubert RE, Geiss LS. Prevalence and incidence of non-

insulin-dependent diabetes. In: Harris MI, Cowie CC, Stern MP, Boyko

EJ, Reiber GE, Bennett PH, eds. Diabetes in America. 2nd ed.Washing-

ton, DC: National Institutes of Health; 1995.

29. Glaser B, Strauss A. Discovery of Grounded Theory. Chicago, Ill: Aldine;

1967.

30. Lyons RF, Sullivan MJL, Ritvo PG, Coyne JC. Relationships in Chronic

Illness and Disability.Thousand Oaks, Calif: Sage Publications; 1995.

31. Trief PM, Sandberg J, Greenberg RP, et al. Describing support: a qual-

itative study of couples living with diabetes. Fam Systems Health.

2003;21:57-67.

32. Sholevar G, Perkel R. Family systems intervention and physical illness.

Gen Hosp Psychiatry. 1990;12:363-372.

33. Olson D. Circumplex model of marital and family systems: assessing

family functioning. In:Walsh F, ed. Normal Family Processes. 2nd ed. New

York, NY: Guilford Press; 1993.

34. Konen J, Summerson J, Dignan M. Family function, stress, and locus of

control. Arch Fam Med. 1993;2:393-402.

35. Olson DH, McCubbin HI, Barnes HL, Larsen AS, Muxen MJ,Wilson

MA. Families:What Makes Them Work. Newbury Park, Calif: Sage Pub-

lication; 1989.

36. Lavee Y, Olson DH. Family types and response to stress. J Marriage Fam.

1991;53:786-798.

37. Gorell DM, Olson DH. Circumplex model of family systems: integrat-

ing ethnic diversity and other social systems. In: Mikesell RH, Luster-

man D, McDaniel SH, eds. Integrating Family Therapy: Handbook of Fam-

ily Psychology and Systems Theory. Washington, DC: American

Psychological Association; 1995.

38. Karpel M. Individuation: from fusion to dialogue. Fam Process.

1976;15:65-82.

39. Franz MJ, Bantle JP, Beebe CA, et al. Diabetes Care. 2002;25:148-198.

Invitation for Letters to the Editor

The Journal of Nutrition Education and Behavior welcomes timely and succinct letters expressing responsiblecriticism or reaction to material published in previous issues and letters calling attention to topics of generalinterest to nutrition education professionals. Letters should be addressed ‘‘To the Editor.’’ The Editor maysend letters to other persons for reaction or rebuttal.

Submission of a letter to the editor constitutes permission for the Journal of Nutrition Education and Behav-ior to publish it in our pages with appropriate editing and abridgment.Authors of letters to the editor mustacknowledge financial and other conflicts of interest within the letter and/or in an author affiliation footnoteto accompany the letter.

Letters should be no more than 200 words.They should be typewritten and double-spaced with 1'' margins.Letters regarding articles published in JNEB will receive priority preference for publishing.

Send letters to:

Sandra K. Shepherd, PhD, RD, Editor Journal of Nutrition Education and Behavior

Department of Food Science and Human NutritionColorado State University

502 W. Lake Street, Campus Delivery 1571Fort Collins, CO 80523-1571

Tel: (970) 491-3060; Fax: (970) 491-7252; e-mail: managingeditor@jneb.org