13-2-66.pdf - American Diabetes Association

Health-Care Communicationand Selected PsychosocialCorrelates of Adherencein Diabetes Management

Lynda A. Anderson, PhD

Meeting the needs of older adults with diabetes mellitusrequires understanding the multiple factors that interactto influence patient adherence to the prescribedregimen and adjustment to the disease. Herein, wepropose a conceptual framework to explain patientadherence and to discuss practical strategies forenhancing adherence. The conceptual frameworkis based on the cycle of care and is expanded toincorporate major theories in health-behavior change.Based on this review, numerous factors were identifiedthat can affect patient adherence. First, features of theregimen, such as complexity, duration, and extent ofbehavioral change, can negatively influence patientadherence. Second, psychosocial and environmentalbarriers to patient adherence are described. A majorarea of importance is patient-provider relationships.Specifically, patients' and professionals' attitudes,beliefs, communication processes, and subsequentreactions to the consultation can affect adherence.Suggestions are made regarding strategies to enhanceadherence, including the need to assess and monitorthe patient's beliefs and practices, enhance thecommunicative skills of health-care professionals andpatients, and develop programs specifically targeted atpromoting professional and patient problem-solvingskills. The problem of generalizing from currentresearch is that few empirical studies have focused ondiabetes management among older adults. DiabetesCare 13 (Suppl. 2):66-76, 1990

The increased challenges of managing diabetesmellitus among older adults are highlighted bythe expanding numbers of elderly people con-fronted with diabetes late in life (1,2). Diabetes

regimens often demand that patients perform multiplecomplex treatments over extended periods. The effec-

tiveness of such treatments depends on the efficacy ofthe treatment prescribed and the extent to which pa-tients are able to initiate and sustain the treatments thatmake up the regimen. Meeting the needs of older adultswith diabetes requires understanding the multiple fac-tors that interact to determine the efficacy of the pre-scribed treatment and attending to the factors that facil-itate or inhibit patients' abilities to master the treatmentregimen and successfully cope with the disease.

Herein, we review the current state of knowledge re-garding major issues in diabetes management amongolder adults. First, a conceptual framework is providedto facilitate the organization of the literature review.Second, a review of the diabetes-care literature is pre-sented. In view of the dearth of empirical data on olderpeople with diabetes, in many instances it is necessaryto rely on findings from the broader literature on adultswith diabetes and other chronic illnesses. Finally, anoverview of strategies used to enhance knowledge andskills in treatment adherence and directions for futureresearch are described.

CONCEPTUAL FRAMEWORK

As a framework for this review, a conceptual model isprovided for understanding the possible linkages amongfactors related to self-management behaviors (Fig. 1).This framework is based on a prevailing model of health

From the Department of Health Behavior and Health Education, School of PublicHealth, University of Michigan, and the Geriatric Research, Education, andClinical Center, Veterans Administration Medical Center, Ann Arbor, Michigan.

Address correspondence and reprint requests to Lynda A. Anderson, PhD,Department of Health Behavior and Health Education, School of Public Health(ID-M5047, University of Michigan, Ann Arbor, Ml 48109-2029.

66 DIABETES CARE, VOL. 13, SUPPL. 2, FEBRUARY 1990

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/13/Supplem

ent_2/66/508034/13-2-66.pdf by guest on 16 August 2022

L.A. ANDERSON

ANTECEDENTFACTORS• Preferences• Expectancies• Situations!• Personal \

\

CONSULTATIONPROCESSES• Informational Exchanges* Affect

IMMEDIATEOUTCOMES• Recall/Comprehension• Satisfaction with Care• Reduction in Concerns

FACILITATING/INHIBITING

FACTORS• Situational Demands• Health Beliefs• Social Support

LONG-TERMOUTCOMES• Health Status• Functioning• Mortality

INTERMEDIATEOUTCOMES• Adherence

FIG. 1. Overview of health-care communication and selected psychosocial variables related to adherence. Solid linesdescribe relationships among major components. Modified from Pendleton et al. (4).

care, the cycle of care (3,4), and leading theories inpredicting and explaining adherence behavior (5,6).Within this framework, two primary sets of factors arebelieved to impact on treatment adherence: 7) factorswithin the health-care system, involving health-careprovider-patient interactions and their immediate out-comes, and 2) factors within the patient's interpersonaland social environment, including psychological andsocial elements that facilitate or inhibit adherence to theprescribed treatment regimen. These major componentsare conceived of as operating jointly or interactively toinfluence patient adherence and treatment outcomes.

As indicated in Fig. 1, factors related to the health-care system are depicted as an antecedent-process-out-come relationship. For example, the patient's and prov-ider's beliefs (antecedent factors) are believed to impacton the interaction during the consultation (process) andaffect the patient's subsequent reaction to the consul-tation (4). The patient's reaction to the consultation af-

fects subsequent adherence to the prescribed regimen.Interpersonal factors and aspects of the patient's

broader social environment also inhibit or facilitate ad-herence (Fig. 1). These factors include 7) the patient'sbeliefs about the disease and the prescribed efficacy ofthe regimen and 2) the patient's social environment. Inturn, inhibiting and facilitating factors impact on sub-sequent medical interactions and are believed to affectlong-term treatment outcomes. This framework is usedto underscore the fact that adherence is not a static con-cept and needs to be placed within a broader contextinvolving the patient, the patient's social network, andthe health-care system.

LITERATURE REVIEW

There is no need to dwell on the well-documented factthat people of all ages make many errors on both tests

DIABETES CARE, VOL. 13, SUPPL. 2, FEBRUARY 1990 67

Dow



SELECTED CORRELATES OF ADHERENCE

of knowledge and performance skills in diabetes man-agement. Many excellent reviews are available con-cerning patient performance errors and adherenceproblems in diabetes management (7-9). Estimates ofnonadherence and performance errors range from 33 to82% depending on the population and the behavior un-der study (9-12). Several important lessons can be drawnfrom this literature and are briefly summarized below.

Treatment programs vary in terms of complexity ofimplementation, duration, and extent of behavioralchange required (13). These features of the regimen canaffect patient adherence. First, complexity of the regi-men has been linked to increased patient error and greaterdifficulties in efforts to achieve the established goals(14,15). Therefore, practitioners must take clinically ap-propriate steps to reduce the complexity of the regimensthey prescribe. This may be particularly critical for olderadults who are prescribed a number of different regi-mens for various health conditions. It is important thattreatment be tailored to meet the life-style of the personinvolved, because most treatments require personal self-management behaviors by the individual or significantother(s) (14). Tailoring involves fitting the treatment reg-imen to the habits and rituals in the individual's dailylife (15). Second, the period over which the behaviormust be performed is negatively related to adherence.Several investigations have shown that performance skillsdecline over time, particularly with insulin injections(10,16,17). Even after patients acquire a specific skill,their performance should be regularly monitored andreinforced. Third, the degree to which the person is re-quired to make changes in his/her daily routine is neg-atively related to adherence. Substantial evidence in-dicates that there are considerable variabilities in ratesof adherence with different aspects of the regimen(12,16,18). In general, patients are more adherent tomedications than to life-style alterations such as dietaryand activity programs (19-21). Recommendations thatdo not restrict or alter existing habits and routines aremore likely to be followed. When significant behavioralchanges are required, as in the case of many of thedietary and activity components of the regimen, moretime and greater efforts are required.

There is no clear-cut evidence to suggest that age iscorrelated with adherence (14). As numerous studiesdemonstrate (22,23), adherence is largely unaffected byage when the complexity of the regimen is properly con-sidered. Of course, this does not imply that age is anirrelevant factor in adherence, only that knowing an in-dividual's age provides almost no useful information re-garding the extent of adherence that can be anticipated.

In addition, there are few available data on which todetermine whether the magnitude of performance errorsare really different for younger and older adults withdiabetes mellitus. To date, the available findings areconflicting (24,25). Older adults have not been includedin many of these investigations, even though they mayin fact have more complex treatment regimens or func-tional limitations that increase management problems.

Caution is warranted in generalizing these findings toolder adults, because much of our knowledge of dia-betes management comes from highly select samples ofyounger respondents. As a consequence, there is an ur-gent need for studies of diabetes self-management amongolder adults.

A major shortcoming of the compliance literature isthat psychosocial and environmental barriers to diabetesmanagement were often overlooked. Thus, two majorgoals were adopted for this review. First, we reviewedthe following key issues pertaining to diabetes manage-ment among older adults: 7) the effect of health-careprovider-patient interaction on adherence, 2) how anindividual's perceptions and coping strategies assist him/her in adjusting to the disease and treatment regimen,and 3) how the patient's social network affects adher-ence. Second, based on the literature, practical sugges-tions involving intervention strategies are discussed, andsuggestions for future research are described.

FEATURES OF PATIENT-PROVIDER RELATIONSHIPS

The quality of the patient-provider relationship is an im-portant factor in achieving adherence to medical advice.Health-care provider—patient interactions are complexprocesses that involve preferences and beliefs that par-ticipants bring to the interaction, the degree of com-municative skill the participants can draw on during theinteraction, and their reaction to the consultation.Antecedent factors. There is increasing recognition thatthe success of patient-provider communication is due,in part, to preexisting factors brought to the interactionby participants (26). Accordingly, these factors areviewed as critical determinants of communication pro-cesses and have major implications for understandingpatient adherence.

There is substantial evidence that patients have spe-cific desires for information and involvement in the clin-ical interactions (27-32). Regardless of age, the majorityof patients express a desire for information about theirillness and treatment program (27-29). Moreover, pa-tients report more dissatisfaction with the informationgiven to them by physicians in the course of the med-ical visit than with any other aspect of their medicalcare (3).

There is considerable variability in desire for involve-ment in health-care processes, with the exception ofage/cohort differences. Older adults, in general, expressless desire for involvement than younger adults (27-32).One potential explanation is that older adults grew upin an era when more authority was attributed to health-care professionals (30). A second hypothesis is that be-cause older adults are often confronted with more com-plex health problems than younger adults, older adultschoose to play a less active role and may prefer clini-cians to make the major decisions. Interestingly, the pa-tient's desire for personal control in clinical interactionsis significantly related to the duration of diabetes and tosatisfaction with care (33). These findings suggest thatamong a population of older men with diabetes, those


Dow



L.A. ANDERSON

who desire personal control have had diabetes for alonger period and may be less satisfied with what tran-spired during the medical visit than those who do notdesire control. The patients' experiences with an illnessand trust in their physician may influence their desiredrole in the interaction and their subsequent responses tomedical recommendations. These are questions that meritfurther investigation.

Patients' expectations are also viewed as importantdeterminants of consultation processes (34-36). Expec-tations refer to what an individual anticipates will hap-pen (37). Although the majority of studies have focusedon interactions in acute-care situations, these studiesindicate that when patients' expectations about their careare consistent with their perceptions of what actuallytranspires, they are more satisfied and more adherent tothe prescribed therapy (35,36). Unfortunately, patients'expectations often go unrecognized by health-care pro-fessionals (35).

The effects of stereotyping by health-care providersare widely documented (38), and there is considerableevidence that stereotypes about the elderly are partic-ularly salient and prevalent (39,40). For example, health-care providers have been found to be less responsive topsychosocial issues raised by older patients (41) andhave been noted to diminish the significance of somesymptoms by ascribing them to aging (42). Researchreveals that when providers interact with patients whofall into a stereotyped group (e.g., women orolder adults),the providers' behavior may elicit behavior from thepatient that confirms the stereotype (43,44). As re-searchers concerned with stereotyping suggest, biasesheld toward patients may be more subtle than attitudemeasures may reflect and may be communicated to pa-tients through alterations in speech and language pat-terns (45).

Recently, attention has been directed toward assess-ing health-care professionals' attitudes toward workingwith patients with diabetes (46). A recent study by Lauxet al. (47) demonstrated that the degree to which phy-sicians desired to work with these patients correlatedwith the perceived manageability of the disease. Addi-tionally, physicians indicated that they did not expectpatients to succeed at managing their disease with dietand exercise. These findings have negative connotationsfor patient adherence in that clinicians' expectations maybe communicated to patients and impede their effortsto adhere to the regimen.

As noted by Mechanic (48), the utility of knowingpatients' or physicians' perceptions may depend on thedegree of agreement regarding their interpretations ofthe situation. It is suggested that the extent to whichpatients and providers hold congruent expectations willresult in more satisfied patients and more adherence tothe prescribed regimen. A number of studies have shownthat, more often than not, patients and physicians holddissimilar perceptions (49-51). When patients' percep-tions of their diabetes are in conflict with what they aretold by their physician, they express considerable un-

certainty and anxiety about the illness and its manage-ment (52). Furthermore, patients and physicians oftendisagree about the reasons for nonadherence. Pendletonet al. (51) demonstrated that physicians often attributepatients' nonadherence to lack of motivation, whereaspatients attribute nonadherence to situational and phys-iological barriers.

Detection of unmet expectations or conflicting per-ceptions requires an awareness by health-care profes-sionals, and negotiations may be required to obtain amutually satisfactory relationship. If the negotiations endto the satisfaction of all, a desire to maintain and pro-mote the relationship develops. If either party is not sat-isfied with the result, the relationship becomes strained,and termination or dropping out of treatment becomeslikely (53).Communication processes. Examination of commu-nication processes between patients and health-careproviders reveals important information about the roles,content, and goals of the interactions. Numerous studieshave examined the relationship between patient-provi-der communication and patient-care outcomes (see refs.3,54,55 for recent reviews), although few investigationshave directly focused on diabetes care and manage-ment. Communication studies have emphasized thesharing of information between practitioners and pa-tients and the affective elements of the communication.

Many studies have documented the difficulties relatedto information provision from physicians to patients. Thework by Ley and associates (56,57) demonstrated thatthe clarity of the message and the timing of its deliveryare positively related to the patient's comprehension andrecall of information. Ley (57) advises that practitionersshould provide information early in the interaction, em-phasize the importance of their advice, make explicitsuggestions regarding the recommended changes, andprovide directions that are specific and concrete. Onestudy on patient-provider communication in diabetes caredemonstrated that a high frequency of clinicians' state-ments that included providing regimen instructions,sharing medical data, and justifying medical advice weresignificantly related to patients' comprehension of theirdiabetes regimen (58).

Although research to date has relied extensively onfrequency measures (i.e., how often a particular behav-ior occurs), other measures such as duration have beensuccessfully used in describing important elements ofmedical interactions. For instance, several investigatorshave found that the amount of time spent with patientsproviding information is related to better recall (59) andhigher levels of patient satisfaction (60,61). This re-search suggests that the amount of information providedto patients and the justification of medical advice areimportant components of communication and are re-lated to patient-care outcomes.

Advocates of patient consumerism have pointed to theimportance of patient involvement in medical interac-tions and have encouraged more active participation bypatients (30). Although not all researchers agree, when


Dow




patients are encouraged to be more involved, they tendto be more satisfied with their medical care (60). As aresult, a number of recent interventions have focusedon coaching patients to be more assertive through askingmore questions (62) or disclosing information about theirown needs (63). Greenfield et al. (64) revealed the ef-ficacy of an intervention to enhance patient involvementin clinical interactions. A randomized trial was con-ducted to test the efficacy of an intervention to encour-age patients to ask questions and negotiate medical de-cisions with the physician. Analysis of audiotaperecordings of the medical visit revealed that patients inthe experimental group were twice as effective as thosein a control group in eliciting information from the phy-sician. Although no differences were found for patientsatisfaction or knowledge of disease, patients in the ex-perimental group reported a decrease in functional lim-itations.

The special importance of communication betweenhealth-care professionals and older patients has beennoted (65). The effects of patient age on actual com-munication exchange is unclear, although there are sev-eral investigations that suggest the importance of thisfactor. For example, Greene et al. (41) found no differ-ence between physicians' speech (e.g., frequency ofquestions) to younger and older patients. However, whenphysicians' communication was rated by a panel ofjudges, their behavior was judged to be less supportive,egalitarian, and respectful to older patients than youngerpatients. Furthermore, older patients were judged to beless successful in capturing and maintaining the atten-tion of physicians. Undoubtedly, there is a pressing needfor research on interaction involving communicationbetween elderly patients and health-care professionals.

The necessity of responding to patients on an affectiveas well as a cognitive level has been documented (66,67).Health-care professionals must not only consider whatthey tell their patients but also how information is pre-sented. Although some studies suggest that physicianswho are "warm and friendly" are preferred (3), othersindicate that physicians who are "serious and intense"have more satisfied patients (42). Interestingly, patientswith nonserious and acute ailments prefer physicianswho are perceived to be warm and friendly, whereaspatients with chronic or serious illnesses prefer physi-cians who appear to be more serious and intense (67,68).In general, patients appear to be most consistently sat-isfied when interacting with physicians who are per-ceived as being involved in the interaction and whoappear to be concerned about them (69). One meansthrough which concern and interest may be conveyedto patients is the affective component of the interaction.Summary and practical implications. Many questionsabout patient-provider relationships remain unan-swered; however, it is still possible to design inter-vention strategies to strengthen these relationships.Health-care professionals should apply the principles ofeffective persuasion (70,71). The message must not onlybe well specified but also organized and delivered in a

manner that will allow the patient to attend to and pro-cess it completely (57). Both oral review and writteninstructions are important to reinforce essential points(72). Selection of the type of information to provide shouldbe based on an assessment of the needs and prioritiesof the patient. As indicated by Svarstad (73), carefulattention should be paid to the quality rather than thesheer quantity of information given. Health-care profes-sionals should emphasize the necessity of adherence toparticularly critical aspects of the treatment (i.e., prior-itize the regimen). This is particularly important whenan interdisciplinary team is used, where several healthprofessionals may be instructing a patient.

Although receipt, comprehension, and retention ofinformation about the treatment regimen are essentialfor adherence, they may not be sufficient (71). In ad-dition, various other factors beyond information provi-sion can enhance information acceptance and promoteadherence to the prescribed regimen. There is a needfor improved understanding about the patient's interpre-tation of the information provided and the way it is in-corporated into the individual's own "explanatory modelof his/her illnesses" (74).

Health-care professionals and patients must work to-gether to find solutions to management problems. Health-care providers must be approachable and should allowpatients to be involved in the interaction. As indicatedpreviously, strategies such as record review and coun-seling (64) and modeling (63) may assist in promotingactive patient involvement. Other techniques, such aspatient-provider contracts, that involve the patient intherapeutic decisions and in setting treatment goals maybe useful for achieving these objectives (75). However,some patients may respond better to some techniquesthan others (73).

Finally, patients should be encouraged to be success-ful with self-management programs. Self-monitoring canprovide information about responses to treatment at homethat can be important in making decisions about theregimen. When clinically appropriate, patients may needto start with a simplified regimen to which more com-plex behaviors are gradually added to enhance theirchances of early success and promote a feeling of self-efficacy. Both practitioners and patients must establishrealistic expectations for successful self-management ef-forts, including the identification of short- and long-termgoals that can be verified through self-monitoring andpatient assessment (76).

FACILITATING AND INHIBITING FACTORS

During the last decade, there has been increasing inter-est in the role that patients' perceptions of their illnessand regimen play in predicting adherence. The mostwidely applied model in diabetes management has beenthe health belief model. The health belief model pos-tulates that perceived susceptibility and seriousness(threat), benefits and barriers (value), and cues to actionare major determinants of health behavior (5). Beckerand Janz (77) recently summarized the findings from


Dow



L A ANDERSON

health belief model research in diabetes management.Their review provides support for consistent yet mod-estly sized correlations between some or all of the majorcomponents of the health belief model and treatmentadherence.

The health belief model is still used as a major frame-work for predicting adherence in diabetes management.More recent research has attended more closely to dis-ease factors that may influence management and hasadded various social factors to the model. Jenny (78)found that patients using insulin indicated stronger be-liefs in the benefits of the regimen, more concerns abouttheir diabetes, and better adherence to the regimen thanthose not using insulin. In a subsequent report examin-ing age effects among the same sample, the oldest pa-tients had lower scores on the perceived value of themedication and exercise compared with younger re-spondents (79). Many of the features of the regimen,such as insulin use, were highly correlated with age.Similarly, Davis et al. (80) documented the importanceof attending to the heterogeneity of treatment regimensand psychosocial factors among patients with diabetes.They indicated that spurious results may be producedby inadequate analysis of heterogeneous groups of pa-tients with diabetes.

There are recent data demonstrating a relationship be-tween psychosocial issues and long-term survival of pa-tients with diabetes. Davis et al. (81) examined the re-lationship between patient survival and a set of predictorvariables, including demographic, clinical, psychoso-cial, behavioral, and physiological covariates. The ran-domly selected sample included 343 community-basedpatients with non-insulin-dependent diabetes mellitus(NIDDM) who entered the study in 1981-82 and werereexamined in 1985-86. The results indicated that thevariables most associated with the risk of mortality wereage, social impact of diabetes, complexity of diet regi-men, history of smoking, and renal function. The psy-chosocial variables, social impact and complexity of dietregimen, were obtained from the Diabetes EducationProfile, which was completed on entry to the study.Even when demographic, physiologic, and diabetestreatment outcomes were included in the predictionequation, an association was found between psycho-social variables and the risk of mortality. As recom-mended by the authors, further research is needed todetermine whether patients' perceptions of the socialimpact of diabetes and ratings of the complexity of thediet regimen are amenable to change through interven-tions designed to enhance social support and problem-solving skills.

Other theoretical models have the potential to makeimportant contributions to understanding diabetes man-agement. Research on diabetes management may ben-efit from research based on the self-regulatory model ofillness behavior of Leventhal et al. (6) and social learn-ing theory (76). With their model, Leventhal et al. at-tempt to explain how people construct an explanatorymodel of their illness (i.e., identify the problem and its

causes and consequences) and develop plans to copewith it. As a result, behavior such as adherence to healthrecommendations is viewed as dependent on the indi-vidual's cognitive representation of his/her current healthstatus, plans for altering the current state, and tech-niques or rules for appraising progress. Self-regulationof diabetes depends in part on common sense modelsof symptoms and blood glucose fluctuations.

Although not directly applied to adherence behavior,several recent investigations have examined symptomawareness and blood glucose estimation in adults withNIDDM (82,83). Diamond et al. (83) found that among52 adults with difficult-to-control NIDDM, most of thepatients believed they could detect hyperglycemia, eventhough patients' estimates of their blood glucose levelswere determined to be far from accurate. Of interest tofuture research is the extent to which patients use sub-jective symptoms in diabetes management, how subjec-tive symptom perceptions and blood glucose estima-tions may inhibit or facilitate adherence, and howsubjective symptom use possibly interacts with moreformal programs of blood glucose monitoring. Anotherimportant application of this model is to examine howolder adults perceive diabetes in relation to the presenceof other chronic illnesses that also demand time andattention. This is an area of critical importance amongolder adults confronted with multiple chronic illnesses.

Taken together, this research indicates that health-careprofessionals should ascertain and respond to patients'health beliefs. In particular, knowledge of patients' be-liefs may help to ensure that they are provided with theappropriate type of information. An assessment of theirbeliefs can provide an important baseline for futurecomparison and a means for taking into account adher-ence as a part of the ongoing process of care.Social support. Social factors are clearly implicated inadherence to medical recommendations. Becker andMaiman (84) revealed a growing body of evidence thatsocial support, particularly from patients' families, playsan important role in initiating and maintaining adher-ence. A major misconception about older adults is thatthey are alone and isolated. The available data do notsupport this common assumption (85). Indeed, the bulkof the data suggest that older adults are surrounded bya support network of family and friends.

Research on social support among adult populationswith diabetes mellitus is scarce, but there are severalstudies that attest to its importance. Among a sample ofmen with diabetes, Edelstein and Linn (86) found thatmen who perceive their environment to be high inachievement orientation were in better metabolic con-trol. Furthermore, patients' perceptions of family func-tioning have been shown to be related to good meta-bolic control (87). However, the relationship betweensocial support and control is not always positive orstraightforward. For example, Heitzmann and Kaplan(88) demonstrated that men who expressed greater sat-isfaction with their social network were found to be inpoor glycemic control, whereas women who expressed


Dow




greater satisfaction were found to be in good control.Thus, there is much work left to be done on social sup-port and diabetes management. Note that these studiesexamined treatment outcomes rather than adherencebehaviors. In addition, cross-sectional studies are prob-lematic in research on social support, because the pres-ence of supportive others is often confounded with aperson's ability to elicit or make use of the support thatis available (89). Future research should examine howsocial network composition and different types of sup-port may affect diabetes management.Summary and practical implications. Relationshipsamong interpersonal factors, social variables, and ad-herence behaviors (and mortality) have been presented.Historically, the early search for correlates of adherencewas descriptive rather than explanatory, and even whenderived from theoretical propositions, the approach wassometimes fragmented. One problem has been thattreatment adherence, a behavioral variable, is not al-ways included in diabetes research. Moreover, clinicalindices of metabolic control are often treated as theequivalent of measures of compliance. Such practicesare inappropriate and misleading because metaboliccontrol can be influenced by a great number of otherfactors in addition to adherence to the prescribed treat-ment (90). In future research, more care should be takenin differentiating between these constructs.

With these limitations in mind, evidence from re-search on health beliefs and social support is highlysuggestive of strategies for intervention. It should be ap-parent that practitioners must have an understanding ofthe situational demands, the patient's beliefs about theillness and its treatment, and the patient's social envi-ronment to provide optimal care. More integrated modelsthat take these factors into account are required to ad-vance our understanding of diabetes management in olderadults. A needs assessment can help to identify relevantbeliefs, behaviors, and perceived social support systemsthat may influence the consultation process and treat-ment adherence. Future research should examine thereliability and validity of instruments for use with olderadults with diabetes.

Although this paper focuses largely on treatment ad-herence, it should be kept in mind that living with di-abetes can impose enormous challenges on both theindividual and significant other(s) far beyond their at-tempts to manage the prescribed therapy. Patients' ad-justment to diabetes and their overall quality of life areimportant areas in need of additional research. Diabetesand its complications may result in considerable de-mands on a person's coping resources. For this reason,living with diabetes is an important outcome affectingthe collective health and well-being of patients with di-abetes.

PRACTICAL IMPLICATIONS

The cycle of care provides an important framework forexamining various factors relating to patient-care out-comes. Treatment adherence is viewed as a process ratherthan a stationary end point of care. In essence, inter-

personal relationships are the key to the adherence phe-nomenon. Moreover, any intervention or influence shouldbe viewed as reciprocal and demands the involvementof patients, social networks, and health-care profession-als.

With regard to the major components described aspart of the framework in Fig. 1, there are several generalconclusions. These are organized around strategies in-volving professional development, program develop-ment for patients and their families, and policy devel-opment.Professional development. Evidence suggests thatpractitioners must learn various aspects of communi-cating with patients and strategies for promoting patientadherence. Factors influencing patient-provider com-munication include patients' and providers' beliefs abouttheir relationship, the nature of the visit, and goals tobe accomplished during the visit. The literature containsnumerous examples of training programs to facilitatecommunication skills of physicians (91). These pro-grams primarily target techniques such as gatheringmedical data, and few attend to patient-education strat-egies. Practitioners must learn effective patient-educa-tion strategies to use during their limited time with pa-tients.

Practitioners must attend to the situational, cognitive,and behavioral aspects of diabetes management that mayfacilitate or inhibit treatment adherence. One practicalapproach to the problem is for practitioners to assesspatient expectations and behaviors at regular intervals.Simonds (92) suggests that a behavioral-educational di-agnosis can be applied to guide this assessment process.This process includes identifying relevant patient be-haviors and practices, collecting information on factorsthat influence behavior, implementing the educationaland behavioral strategies to promote adherence, andmonitoring and evaluating the progress patients maketoward achieving the desired treatment outcomes.Program development. There are numerous interven-tion programs that have been applied to improve patientmanagement and treatment outcomes in diabetes mel-litus. Much of the work is reviewed in two recent pub-lications (93,94). Implications for practice are brieflydiscussed herein. First, there does not appear to be adirect relationship between knowledge of the diseaseand treatment adherence. Attempts to provide infor-mation about diabetes have generally had little effect onadherence behaviors (3). Second, enhanced educa-tional programs (93) intended to educate patients abouttheir treatments, either alone or combined with behav-ioral strategies, have met with moderate success. Fi-nally, a number of strategies to enhance patient skillsand involvement in their care have been introduced.These include self-monitoring programs, contingencycontracting, patient communication enhancement tech-niques, and social support strategies. Self-monitoringstrategies have been shown to be of value in promotingadherence to medications among those using insulin (95).The efficacy of self-monitoring programs for patients notusing insulin remains equivocal (96,97). Major prereq-


Dow



LA. ANDERSON

uisites for self-monitoring interventions include the ac-ceptability of the procedure to patients and their func-tional status (98). Furthermore, research on thecircumstances in which self-monitoring is effective isrequired.

Contracting between the patient and health-careprofessional is an example of a strategy aimed at en-hancing communication (see ref. 75 for a recent re-view). Contracting in diabetes management has met withvaried success (99). Unfortunately, interventions aimedat promoting patient communication have yet to ex-amine patient adherence as one of their outcomes. Fi-nally, social support enhancement is another key areafor intervention (100).

Although various types of strategies appear to have amoderate effect on changing patients' adherence andmetabolic control (93), the specific mechanisms throughwhich these changes occur remain to be identified. Ad-ditionally, the manner in which patients' characteristicsinfluence responses to interventions remains a subjectfor further investigation.

Unfortunately, most intervention studies have by-passed age as a factor for study. From the gerontologicalliterature, we know that older adults are amenable tonumerous intervention programs. As indicated by Engel(101) in his review of behavioral medicine, the appli-cations of behavioral change strategies for older adultsliving in the community do not differ from those neededfor younger adults. However, it is important to recog-nize the heterogeneity of the elderly population and in-clude a full range of age groups when examining inter-vention research. Studies of diabetes management amongolder adults should take into account underlying social,functional, and psychological changes that accompanyage. Considerations for older adults must be expandedto individuals' cognitive and functional abilities to carryout the prescribed treatments and, when appropriate,address how other acute and chronic illnesses may im-pact on diabetes management.Policy development. Examples of the effects of policychanges and developments are most evident from inter-vention programs conducted in long-term-care facilities(102,103). A final area of interest related to diabetescare involves an understanding of the current state ofmedical care reimbursement with relation to older adultswith diabetes. It is important that health-care providersbe informed of how changes in these policies may im-pact on reimbursements for diabetes-related treatments.For example, with the implementation of the MedicareCatastrophic Coverage Act, the Part B drug program willintroduce coverage for prescription drugs and insulin,previously not covered under Part B of Medicare.

CONCLUSIONS AND DIRECTIONS FOR FUTURE RESEARCH

The fact that treatment adherence continues to bea significant problem after more than two decadesof research attests to the complexity of the issue.Information about patient-provider interactions and

the psychological and social correlates of adherence of-

fers important insight into understanding patients' per-ceptions and behaviors related to diabetes management.Research findings are converging that document the im-portance of psychosocial variables to diabetes manage-ment and treatment outcomes. This review of existingdata documents the importance of psychological pro-cesses underlying treatment adherence. However, muchof the work represents associations between various fac-tors and adherence behaviors as they occur in naturalsituations; there is a dearth of experimental studies con-cerning these relationships.

The factors outlined in Fig. 1 and discussed hereinform a framework for examining the perceptions andbehaviors of older adults with diabetes. There is accu-mulating evidence that a clearer understanding of thecognitive processes underlying patients' health actionsshould allow us not only to predict behavior from as-sessment of these variables but also to influence thesevariables through intervention. Future collaborative ef-forts among investigators, in terms of the conceptuali-zation and measurement of these processes, should ac-celerate progress along these lines.

Despite the fact that numerous studies have been con-ducted on diabetes management, a great need for ad-ditional research remains. There is a striking dearth ofempirical work concerning diabetes management amongolder adults. Given the increasing number of older adultswith diabetes and some suggestive evidence that olderand younger patients might react differently to health-care processes, researchers should be aware of the needto explore possible interactions with age in their studies.

There are few studies of patient-provider relationshipsthat are experimental in nature. Future studies shouldinclude analyses of the conversational features of com-munication that can depict patient-provider role rela-tionships, the form and content of the verbal com-munication, and the nonverbal messages that arecommunicated. It is also important that we attend moreclosely to how communication processes influence pa-tient-care outcomes.

Another important research area is the investigationof how patients' perceptions of their disease and thesocial impact of diabetes influence diabetes control sta-tus, health status, functioning, and survival. It is hopedthat the contributions of future research in exploring themanagement of diabetes in late life will be instrumentalin promoting positive health among older adults.

ACKNOWLEDGMENTS

I thank Irwin Rosenstock, PhD, for valuable commentson an earlier draft of this article.

REFERENCES

1. Bennett PH: Diabetes in the elderly: diagnosis and epi-demiology. Geriatrics 39:37-41, 1984

2. Herman WH, Sinnock P, Brenner E, Brimberry JL, Lang-


Dow




ford D, Nakashima A, Sepe SJ, Teutsch SM, Mazze RS:An epidemiologic model for diabetes mellitus: inci-dence, prevalence, and mortality. Diabetes Care 7:367-71, 1984

3. Pendleton D: Doctor-patient communication: a review.In Doctor-Patient Communication. Pendleton D, HaslerJ, Eds. London, Academic, 1983, p. 5-53

4. Pendleton D, Schofield T, Tate P, Havelock P: The Con-sultation: An Approach to Learning and Teaching. Ox-ford, UK, Oxford Univ. Press, 1984

5. Janz NK, Becker MH: The health belief model: a decadelater. Health Educ Q 11:1-47, 1984

6. Leventhal H, Nerenz DR, Steele DJ: Illness representa-tions and coping with health threats. In A Handbook ofPsychology and Health. Baum A, Singer J, Eds. Hillsdale,NJ, Erlbaum, 1984, p. 219-52

7. Rosenstock IM: Understanding and enhancing patientcompliance with diabetic regimens. Diabetes Care 8:610-16, 1985

8. Hamburg DA, Elliott GR, Parron D (Eds.): Health andBehavior: Frontiers of Research in the Biobehavioral Sci-ences. Washington, DC, Natl. Acad., 1982, p. 144-57

9. Surwit RS, Scovern AW, Feinglos MN: The role of be-havior in diabetes care. Diabetes Care 5:337-42, 1982

10. Watkins JD, Willians TF, Martin DA, Hogan MD, An-derson E: A study of diabetic patients at home. Am /Public Health 57:452-59, 1967

11. Cerkoney KAB, Hart LK: The relationship between thehealth belief model and compliance of persons with di-abetes mellitus. Diabetes Care 3:594-98, 1980

12. Glasgow RE, McCaul KD, Scafer LC: Self-care behaviorsand glycemic control in type II diabetes. J Chronic Dis40;399-412, 1987

13. Kirscht JP, Rosenstock IM: Patients' problems in fol-lowing recommendations of health experts. In HealthPsychology—A Handbook. Stone GC, Cohen F, AdlerNE, Eds. San Francisco, CA, Jossey-Bass, 1979, p. 189—215

14. Haynes RB: Determinants of compliance: the diseaseand the mechanisms of treatment. In Compliance in HealthCare. Haynes RB, Taylor DW, Sackett DL, Eds. Balti-more, MD, Johns Hopkins Univ. Press, 1979

15. Sackett DL, Haynes RB, Tugwell P: Clinical Epidemiol-ogy: A Basic Science for Clinical Medicine. Boston, MA,Little, Brown, 1985

16. Watkins JD, Roberts DE, Williams TF, Martin DA, CoyleV: Observation of medication errors made by diabeticpatients in the home. Diabetes 16:882-85, 1967

17. Lawrence PA, Cheely J: Deterioration of diabetic pa-tients' knowledge and management skills as determinedduring outpatient visits. Diabetes Care 3:214-18, 1980

18. Orme CM, Binik YM: Consistency of adherence acrossregimen demands. Health Psychol 8:27-43, 1989

19. Alogna M: Perception of severity of disease and healthlocus of control in compliant and noncompliant diabeticpatients. Diabetic Care 3:533-34, 1980

20. Lockwood D, Frey ML, Gladish NA, Hiss RC: The big-gest problems in diabetes. Diabetes Educ 12:30-33, 1986

21. Daschner BK: Problems perceived by adults in adheringto a prescribed diet. Diabetes Educ 12:113-15, 1986

22. Anderson JL: Patients' recall of information and its re-lation to the nature of the consultation. In Research inPsychology and Medicine. Osborne D, Ed. New York,Academic, 1979

23. German P, Klein L, McPhee S, Smith C: Knowledge of

and compliance with drug regimens in the elderly.) AmCeriatrSoc 30:568-71, 1982

24. Hulka BS, Cassell JC, Kuppa LL, Burdette JA: Commu-nication, compliance and concordance between physi-cians and patients with prescribed medications. Am )Public Health 66:847-53, 1976

25. Miller LV, Goldstein J, Nicolaisen G: Evaluation of pa-tients' knowledge of diabetes self-care. Diabetes Care1:275-80, 1978

26. Wasserman RC, Inui TS: Systematic analysis of clinician-patient interactions: A critique of recent approaches withsuggestions for future research. Med Care 21:279-85,1984

27. Cassileth BR, Zupkis RV, Sutton-Smith K, March V: In-formation and participation preferences among cancerpatients. Ann Intern Med 92:832-36, 1980

28. Strull WM, Lo B, Charles G: Do patients want to partic-ipate in medical decision making? JAMA 252:2990-94,1984

29. Beisecker AE: Aging and the desire for information andinput in medical decisions: patient consumerism inmedical encounters. Gerontologist 28:330-35, 1988

30. Haug M: Doctor patient relationships and the older pa-tient. J Gerontol 34:852-60, 1979

31. Ruzicki DA: Relationship of participation preference andhealth locus of control in diabetes education. DiabetesCare 7-372-77, 1984

32. Woodward NJ, Wallston BS: Age and health care beliefs:self-efficacy as a mediator of low desire for control. Psy-chol Aging 2:3-8, 1987

33. Anderson LA, DeVellis RF, Boyles B, Feussner JR: Pa-tients' perceptions of their clinical interactions: devel-opment of the multidimensional desire for control scales.Health Educ Res 4:383-97, 1989

34. Tinsley HEA, Workman R, Kass R: Factor analysis of thedomain of client expectancies about counseling. / Coun-sel Psychol 27:561-70, 1980

35. Korsch BM, Gozzi EK, Francis V: Gaps in doctor-patientcommunication: doctor-patient interaction and patientsatisfaction. Pediatrics 42:855-71, 1968

36. Freemon B, Negrete VF, Davis M, Korsch BM: Gaps indoctor-patient communication: doctor-patient interac-tion analysis. Pediatr Res 5:298-311, 1971

37. Feather NT: Expectancy-value approaches present statusand future directions. In Expectations and Actions: Ex-pectancy-Value Models in Psychology. Feather NT, Ed.Hillsdale, NJ, Erlbaum, 1982, p. 395-420

38. Blalock SJ, DeVellis BM: Stereotyping: the link betweentheory and practice. Patient Educ Counsel 8:17-25, 1986

39. Rodin J, Langer E: Aging labels: The decline of controland the fall of self-esteem. / Soc Issues 36:12-29, 1980

40. DeVellis BM,. Wallston BS, Wallston KA: Stereotyping:a threat to individualized patient care. In Current Per-spectives in Nursing: Social Issues and Trends. Vol. 2.Flynn MH, Ed. St. Louis, MO, Mosby, 1980

41. Greene MG, Adelman R, Charon R, Hoffman S: Ageismin the medical encounter: an exploratory study of thedoctor-elderly patient relationship. Language Commun6:113-24, 1986

42. Waitzkin H: Doctor-patient communication. Clinicalimplications of social scientific research. JAMA 252:2441-46, 1984

43. Green SK: Attitudes and perceptions about the elderly:current and future perspectives. Int j Aging Hum Dev13:99-119, 1981


Dow



L.A. ANDERSON

44. Snyder M, Tanke ED, Berscheid E: A social perceptionand interpersonal behavior: on the self-fulfilling natureof social stereotypes. / Person Soc Psychol 35:656-66,1977

45. Sebastian RJ, Ryan EB: Speech cues and social evalua-tion: markers of ethnicity, social class, and age. In SocialCognition and Language. Roloff ME, Berger CE, Eds.Beverly Hills, CA, Sage, 1982, p. 112-43

46. Anderson RM, Donnelly MB, Gressard CP, Dedrick RF:Development of the diabetes attitude scale for health-care professionals. Diabetes Care 12:120-27, 1989

47. Laux L, Vallbona C, Merrill J, Baker S, Pavlik V: Physi-cian affect and patient beliefs related to expected failurein the management of diabetes (Abstract). Diabetes37:55A, 1988

48. Mechanic D: The experience and expression of distress:the study of illness behavior and medical utilization. InHandbook of Health, Health Care and the Health Profes-sions. Mechanic D, Ed. New York, Free Press, 1983, p.591-607

49. Rakowski W, Hickey T, Dengiz AN: Congruence of healthand treatment perceptions among older patient andproviders of primary care. Int) Aging Hum Dev 25:67-81, 1987-88

50. Linn MW, Linn BS, Skyler JS, Harris R: The importanceof self-assessed health in patients with diabetes. Dia-betes Care 3:599-606, 1980

51. Pendleton L, House WC, Parker LE: Physicians' and pa-tients' views of problems of compliance with diabetesregimens. Public Health Rep 102:21-26, 1987

52. Mason C: The production and effects of uncertainty withspecial reference to diabetes mellitus. Soc Sci Med21:1329-34, 1985

53. Hayes-Bautista DE: Termination of the patient-practi-tioner relations: divorce, patient style. J Health Soc Be-hav 17:12-21, 1976

54. Hinckley JJ, Craig HK, Anderson LA: Physician-patientinformational exchanges. In Handbook of Language andSocial Psychology. Giles H, Robinson WD, Eds. NewYork, Wiley. In press

55. Roter DL, Hall JA, Littel NR: Patient-physician com-munication: a description of the literature. Patient EducCounsel 12:99-119, 1988

56. Ley P, Whitworth Ma, Skilveck CE, Woodward R, Pin-sent RJHF, Pike LA, Clarkson ME, Clark PB: Improvingdoctor-patient communication in general practice. / RColl Gen Pract 26:720-24, 1976

57. Ley P: Patients' understanding and recall in clinicalcommunication failure. In Doctor-Patient Communica-tion. Pendleton D, Hasler J, Eds. London, Academic,1983, p. 89-108

58. Mazzuca SA, Weinberger M, Kurpius DJ, Froehle TC,Heister M: Clinician communication associated with di-abetic patients' comprehension of their therapeutic reg-imen. Diabetes Care 6:347-50, 1983

59. Bertakis KD: The communication of information fromphysician to patient: a method for increasing patient re-tention and satisfaction. / Fam Pract 5:217-22, 1977

60. Stiles WB, Putnam SM, James SA, Wolf MH: Dimensionsof patient and physician roles in medical screening in-terviews. Soc Sci Med 13:335-41, 1979

61. Bartlett EE, Grayson M, Barker R, Levine DM, GoldenA, Libber S: The effects of physician communicationsskills on patient satisfaction, recall, and adherence. /Chronic Dis 37:755-64, 1984

62. Roter DL: Patient participation in the patient-providerinteraction: the effects of patient question asking on thequality of interaction, satisfaction and compliance. HealthEduc Monographs 5:281-314, 1977

63. Anderson LA, DeVellis BM, DeVellis RF: Effects ofmodeling on patient communication, knowledge, andsatisfaction. Med Care 25:1044-56, 1987

64. Greenfield S, Kaplan S, Ware J, Yan EM, Frank HJL:Patients' participation in medical care: effects on bloodsugar control and quality of life in diabetes.) Gen InternMed 3:448-57, 1988

65. Haug MR, Ory MG: Issues in elderly patient-providerinteractions. Res Aging 9:3-40, 1987

66. Ben-Sira Z: Affective and instrumental components ofthe physician-patient relationship: an additional dimen-sion of interaction theory. ) Health Soc Behav 21:170—80, 1980

67. Hall J, Roter DL, Rand CS: Communication of affectbetween patient and physician. J Health Soc Behav 22:18-30, 1981

68. Milmoe S, Rosenthal R, Blane HT, Chafetz ME, Wolf I:The doctor's voice: postdictor of successful referral ofalcoholic patients. J Abnorm Psychol 72:78-84, 1967

69. Street RL, Wiemann JM: Patient satisfaction with phy-sicians' interpersonal involvement, expressiveness, anddominance. In Communication Yearbook 10. Mc-Laughlin ML, Ed. Newbury Park, CA, Sage, 1987, p.591-612

70. McGuire WJ: The communications-persuasion model andhealth-risk labelling. In Product Labelling and Health Risks.Morris LA, Mariz MB, Barofsky I, Eds. New York, Ban-bury, 1980, p. 99-122

71. McGuire WJ: Attitudes and attitude change. In Hand-book of Social Psychology. Vol. 2. Lindzey G, AronsonE, Eds. New York, Random House, 1985, p. 233-346

72. Becker MH: Patient adherence to prescribed therapies.Med Care 23:539-55, 1985

73. Svarstad B: Physician-patient communication and pa-tient conformity with medical advice. In The Growth ofBureaucratic Medicine. Mechanic D, Ed. New York,Wiley, 1976, p. 220-38

74. Helman CG: Communication in primary care: the roleof patient and practitioner explanatory models. 5oc SciMed 20:923-31, 1985

75. Janz NK, Becker MH, Hartman PE: Contingency con-tracting to enhance patient compliance: a review. Pa-tient Educ Counsel 5:165-78, 1984

76. Bandura A: Social Foundations of Thought and Action:A Social Cognitive Theory. Englewood Cliffs, NJ, Pren-tice-Hall, 1986

77. Becker MH, Janz NK: The health belief model appliedto understanding diabetes regimen compliance. Diabe-tes Educator 11:41-47, 1985

78. Jenny JL: Differences in adaptation to diabetes betweeninsulin-dependent and non-insulin-dependent patients:implications for patient education. Patient Educ Counsel8:39-50, 1986

79. Jenny JL: A comparison of four age groups' adaptationto diabetes. Can ) Public Health 75:237-44, 1984

80. Davis WK, Hess GE, Harrison RV, Hiss RG: Psychoso-cial adjustment to and control of diabetes mellitus: dif-ferences by disease type and treatment. Health Psychol6:1-14, 1987

81. Davis WK, Hess GE, Hiss RG: Psychosocial correlatesof survival in diabetes. Diabetes Care 11:538-45, 1988


Dow




82. O'Connell KA, Hamera EK, Knapp TM, Cassmeyer VL,Eaks GA, Fox MA: Symptom use and self-regulation in 94.type II diabetes. Adv Nursing Sci 6:19-28, 1984

83. Diamond EL, Massey KL, Covey D: Symptom awarenessand blood glucose estimation in diabetic adults. Health 95.Psycho/8:15-26, 1989

84. Becker MH, Maiman LA: Strategies for enhancing pa-tient compliance. / Community Health 6:113-35, 1983 96.

85. Antonucci TC: Personal characteristics, social support,and social behavior. In Handbook of Aging and theSocial Sciences. 2nd ed. Binstock RH, Shanas RH,Eds. New York, Van Nostrand Reinhold, 1985, p. 94- 97.128

86. Edelstein J, Linn MW: The influence of the family oncontrol of diabetes. Soc Sci Med 21:541-44, 1985

87. Cardenas L, Vallbona C, Baker S, Yusim S: Adult onset 98.diabetes mellitus: glycemic control and family function.Am Med Sci 293:28-33, 1987

88. Heitzmann CA, Kaplan RM: Interaction between sex andsocial support in the control of type II diabetes mellitus.y Consult Clin Psychol 52:1087-89, 1984 99.

89. Wallston BS, Alagna SW, DeVellis BM, DeVellis RF: So-cial support and physical health. Health Psychol 2:367-91, 1983 100.

90. Glasgow RE, Wilson W, McCaul KD: Regimen adher-ence: a problematic construct in diabetes research. Di-abetes Care 8:300-301, 1985

91. Carroll JG, Monroe J: Teaching clinical interviewing in 101.the health professions: a review of empirical research.Eval Health Professions 3:21-45, 1980

92. Simonds SK: Individual health counselling and educa- 102.tion: emerging directions from current theory, research,and practice. Patient Counsel Health Educ 4:175-81,1983

93. Padgett D, Mumford E, Hynes M, Carter R: Meta-anal- 103.ysis of the effects of educational and psychosocial in-terventions on management of diabetes mellitus. ) Clin

Epidemiol 41:1007-30, 1988Wing RR: Behavioral strategies for weight reduction inobese type II diabetic patients. Diabetes Care 12:139-44, 1989Skyler JS, Lasky IA, Skyler DL, Robertson EG, Mintz DH:Home blood glucose monitoring as an aid in diabetesmanagement. Diabetes Care 1:150-57, 1978Wing RR, Epstein LH, Nowalk MP, Scott N, Koeske R,Hagg S: Does self-monitoring of blood glucose levelsimprove dietary compliance for obese patients with typeII diabetes? Am J Med 81:830-36, 1986Allen BT, Feussner JR, DeLong ER: Impact of glucoseself-monitoring on type II diabetes mellitus: a random-ized controlled trial comparing blood vs. urine testing(Abstract). Clin Res 35:729A, 1987Southam MA, Dunbar J: Facilitating patient compliancewith medical interventions. In Self-Management ofChronic Disease: A Handbook of Clinical Interventionsand Research. Holroyd A, Creer TL, Eds. Orlando, FL,Academic, 1986, p. 163-87Morgan BS, Littell DH: A closer look at teaching andcontingency contracting with type II diabetes. PatientEduc Counsel 12:145-58, 1988Wilson W, Pratt C: The impact of diabetes educationand peer support upon weight and glycemic control ofelderly persons with noninsulin dependent diabetes mel-litus. Am J Public Health 77:634-35, 1987Engel BT: Behavioral medicine. In Experimental andClinical Interventions in Aging. Walder RF, Cooper RL,Eds. New York, Dekker, 1983, p. 335-42Hamman RF, Michael SL, KeeferSM, Young WF: Impactof policy and procedure changes on hospital days amongdiabetic nursing-home residents—Colorado. MMWR33:621-23, 1984Wylie-Rosett J, Willeneuve M, Mazze R: Professionaleducation in a long-term-care facility: program devel-opment in diabetes. Diabetes Care 8:481-85, 1985


Dow



13-2-66.pdf - American Diabetes Association

Documents