Non-adherence in patients on chronic hemodialysis: an international comparison study

Transcription

1 ORIGINAL ARTICLE JN EPHROL 00( 0000; 00) : Non-adherence in patients on chronic hemodialysis: an international comparison study Christiane Kugler 1, Ilona Maeding 1, Cynthia L. Russell 2 1 Division of Cardiac, Thoracic, Transplantation and Vascular Surgery, Hannover Medical School, Hannover - Germany 2 Sinclair School of Nursing, University of Missouri, Columbia, Missouri - USA Division of Cardiac, Thoracic, Transplantation and Vascular Surgery, Hannover Medical School, Hannover - Germany Ab s t r a c t Background: Adherence to diet and fluid restrictions by adults on hemodialysis treatment is challenging. This study compared non-adherence (NA) to diet and fluid restrictions between adult US and German hemodialysis patients, and assessed potential predictors for NA. Methods: A cross-sectional multicenter comparative study of 456 adult hemodialysis patients was conducted in 12 outpatient-based hemodialysis centers in the United States and Germany. NA was based on selfreport using the Dialysis Diet and Fluid Non-adherence Questionnaire (DDFQ). Laboratory marker, interdialytic weight gain (IDWG) and dialysis adequacy (Kt/V) information were obtained from medical records. Mean time on dialysis was 50 months (range 3-336); mean age was 62 years (range 19-91), with the majority of patients (89.7%) being white. Results: Self-reported frequency of NA to diet was 80.4% and to fluid 75.3% in the total sample. The degree of NA to diet and fluid differed significantly, with the US patients (68.1% vs. 61.1%) reporting less NA when compared with German (81.6% vs. 79.0%) patients (p<0.0001). Phosphorus, albumin, IDWG and Kt/V levels were higher in the US compared with the German subsample (all p<0.0001; IDWG p<0.003). Generalized regression models revealed that education (p<0.01) and smoking (p<0.01) predicted NA to diet, whereas single marital status (p<0.008) and male sex (p<0.04) were independent predictors for NA to fluid. Conclusions: NA persists as one of the most challenging tasks in health care of patients with chronic conditions, including hemodialysis patients. Our findings suggest that patient, condition-related, socioeconomic and health care system related factors may contribute to NA to diet and fluid restrictions. Key words: Diet and fluid, Hemodialysis, Non-adherence In t r o d u c t i o n Chronic hemodialysis is a life-saving procedure for patients with end-stage renal disease (ESRD). The majority of patients use center-based hemodialysis as treatment modality thus offering a unique opportunity to examine adherence behavior in this population. Adherence is defined as the extent to which a person s behavior corresponds with the agreed recommendations of a healthcare provider in terms of taking medications, following a recommended diet and/or executing lifestyle changes (1). To manage the chronic condition successfully, hemodialysis patients should be responsible for many aspects of their own treatment including adherence to medication prescription, adherence to diet and fluid restrictions and complete attendance at hemodialysis sessions (2). Adherence to diet and fluid is paramount for treatment success (2). Failure to adhere may lead to increased complication rates and associated costs, and decreased survival (3-6). According to the World Health Organization (WHO) there are 5 categories of risk factors for patient non-adherence (NA) (1). They include (i) socioeconomic factors, (ii) therapy-related factors, (iii) patient-related factors, (iv) condition-related factors and (v) health care system related factors. In the past 30 years, researchers and clinicians have spent much effort to eliminate relevant risk factors for NA in hemodialysis in the first 4 categories, and to test interventional strategies to support the patient (4, 6-12). However, there is a lack of literature examining the impact of health care system related factors e.g., country effects on outcomes. More specifically, Osterberg and Blaschke (13) suggested in their review on adherence to expand the view that takes into account factors under the patient s control as well as interactions between the patient and the health care system the greatest effect on improving 1

2 Kugler et al: Non-adherence in hemodialysis adherence. This is of equivalent importance for adherence to diet and fluid restrictions in this specific population. However, the comparison of recent findings is limited due to differing NA operational definitions used, large variation in sample sizes and health care system related factors. Table I delineates details from the literature. Accordingly, the overall prevalence of NA to diet using laboratory markers ranged between 2% and 77% (4, 6, 7, 9, 11, 12). The prevalence of self-reported NA was 81.4% (15, 18). However, when examining differences between countries, reported prevalence of NA to diet based on laboratory markers such as phosphorus and albumin, was 2%-77% for US (4, 6, 7, 14, 21) and 4.2%-31% for European patients (9, 17, 22). The prevalence of NA to fluid restrictions based on the interdialytic weight gain (IDWG) was 9.7% to 70% (4, 6, 7, 9, 14, 17, 21-23). Taking the health care system into account, prevalence to fluid NA using IDWG ranged between 9.7% and 49.5% for US (4, 6, 7, 14, 21) and 9.8% and 70% for European patients (9, 17, 22, 23). Self-reported NA to fluid restrictions revealed a 72%-74.6% range (15, 18). These data suggest a wide variation of NA prevalence rates between US and European hemodialysis patients. Thus, the purpose of this study was (i) to compare NA to diet and fluid restrictions between adult US and European hemodialysis patients and (ii) to assess potential predictors for NA. Subjects and methods Design and setting This is a secondary data analysis of a cross-sectional multicenter comparative study which was conducted in 12 outpatient-based hemodialysis centers in the United States (6 centers in the state of Missouri) and in Europe (6 centers in Germany). Data collection procedures were performed under the same methodologically directive conditions using the same instruments for the respective languages after formal forward and backward translation had been performed (24). The study received joint permission by the respective local institutional review boards. Human subjects privacy protection was maintained at all times, with informed consent obtained from all participants. In the United States, a sample of 149 potential participants was selected from the available pool. Of the 149, 11% (n=17) had died or did not meet eligibility criteria. Of the remaining 132 eligible potential participants, 113 participants agreed to participate in the study representing an 86% consent rate. In Germany, 346 subjects were available. Three patients were not willing to participate, which led to a 99.1% consent rate. A native, independent and trained research assistant performed the data collection in each country. Inclusion criteria were sufficient language skills to read and answer the questions on a survey, absence of neurocognitive impairments and being on hemodialysis treatment for at least 6 months. Age of consent was 21 years or older for the US subsample, and 18 years or older for the German subsample. Sample The total sample consisted of 456 hemodialysis patients (n=113 US; n=343 German). Median time on dialysis was 36 months (range months). Median age was 65, ranging from 19 to 91 years with the majority being Caucasians (89.7%). The first column in Table II describes other relevant sample characteristics. Instruments The Dialysis Diet and Fluid Non-adherence Questionnaire (DDFQ) was used for the collection of NA data. This self-report instrument was designed for use in clinical practice and validated by Vlaminck and associates (15). The DDFQ measured NA in adult hospital-based hemodialysis patients in the country of Flanders (Belgium) and was formally translated and validated in 3 other languages: English, German and Turkish (18, 20, 21). The 4-item instrument captures frequency of NA behavior with diet and fluid restrictions by asking the respondent, How many times in the last 14 days did you not follow your diet/fluid guidelines? Responses can range from 0 to 14 days. Degree of deviation is captured by asking To what degree did you deviate from your diet/fluid guidelines? Patients are asked to report deviations from the recommended regimen on a 5-point Likert scale from no deviation to very severe deviation. Biochemical markers collected from the participant s medical records included serum phosphorus (mg/dl) and albumin (mg/dl) to measure diet NA. The biological marker of IDWG (kg) and the dialysis adequacy value of Kt/V, also collected from the participant s medical records, were used as measures of fluid NA. According to the National Kidney Foundation s Kidney Disease Outcomes Quality Initiative (KDOQI) (25), these parameters represent established markers to assess diet and fluid NA. Further details are given by Vlaminck and associates (15). 2

3 JNEPHROL 0000; 00( 00) : TABLE I PREVALENCE REPORTS OF NON-ADHERENCE TO DIET AND FLUID RESTRICTIONS Author, year of publication Country of origin Operational definition Sample/ethnic characteristics Assessment based on (% NA diet) Assessment based on (% NA fluid) Bame et al, 1993 (6) USA Non-compliance 1,230 Potassium (2%) Phosphorus (50%) IDWG (49.5%) Christensen et al, 1995 (14) USA Non-adherence 57 IDWG (42%) Leggat et al, 1998 (7) USA Non-compliance 6,251 Phosphorus (22.1%) IDWG (9.7%) Vlaminck et al, 2001 (15) Belgium Non-adherence 564 SR (81.4%) SR (72%) Casey et al, 2002 (16) United Kingdom Non-compliance 21 IDWG (60%) Durose et al, 2004 (17) United Kingdom Non-compliance 82 Potassium (4%) Phosphorus (31%) IDWG (23%) Hecking et al, 2004 (9) France, Germany, Italy, Spain, Non-adherence 3,039 Potassium (11.7%) Phosphorus (23.8%) IDWG (9.8%) United Kingdom Kugler et al, 2005 (18) Germany and Belgium Non-adherence 916 SR (81.4%) SR (74.6%) Unruh et al, 2005 (4) USA Non-adherence 739 white = 62.5%; black = 32.2%; other = 5.3% Phosphate: whites (60.6%); blacks (34.1%); others (5.3%) Potassium: whites (73.8%); blacks (20.6%); others (5.7%) Barnett et al, 2007 (19) Australia Non-compliance 26 IDWG (53%) Kara et al, 2007 (20) Turkey Non-adherence 160 SR (58.1%) SR (68.1%) Russell et al, 2008 (21) USA Non-adherence 107 Phosphate (68%) Albumin (77%) IDWG (40%) O Connor et al, 2008 (22) United Kingdom Non-adherence 73 Potassium (16%) IDWG (70%) Lindberg et al, 2009 (23) Sweden Non-adherence 4,498 IDWG (30%) IDWG = interdialytic weight gain; NA = non-adherence; SR = self-report. 3

4 Kugler et al: Non-adherence in hemodialysis TABLE II COMPARISON OF SAMPLE CHARACTERISTICS Total (n=456) US subsample* German subsample* Comparison (n=113) (n=343) Coefficient p Value Time on dialysis, mean months ± SD 50 ± ± ± Age in years, mean ± SD 62 ± ± ± Sex (%) Male Female Family status, % Single Married/living with a partner Divorced/widowed Educational level, % Junior high school High school Some college College/ university Ethnicity (%) White African American Asian/ other Comorbid conditions, % Diabetes mellitus Hypertension Peripheral vascular disease Physical impairments Psychosocial problems Smoking, % yes *Total less than 100% refers to missing data. 4

5 JNEPHROL 0000; 00( 00) : Fig. 1 - Comparison of degree of nonadherence (NA) to diet between the US and German subsamples. Data analysis Data were entered and analyzed using SPSS statistical software for Windows (Version 17.0, SPSS Inc., Chicago, IL). Descriptive statistics included percentages for categorical variables, and medians, means, standard deviations and ranges for continuous variables. Univariate analyses of categorical variables were performed by chi-square test or Fisher exact test. Student s t-test and analysis of variance (ANOVA) were used to evaluate the differences between means by adherence cohort. Differences between the US and German subsamples were evaluated using the Mann-Whitney U-test. Generalized linear regression models (GLR) were used to identify diet and fluid NA predictive variables. Variables were included into the model when they had been identified in bivariate analysis to impact adherence. All reported p values are 2-sided. For all analyses, p values of <0.05 were considered statistically significant. Re s u l t s Comparison of subsample characteristics Table II shows sample characteristics by country for hemodialysis patients participating in this study. Median age of the US sample was 59 (range 20-85) which was significantly younger than the German sample (median 65, range 19-91; p<0.020). Comparison of non-adherence to diet between the US and German subsamples Median self-reported diet NA frequency for the entire sample was 3 times, ranging between 1 and 14 times (mean 5.41 ± 5.30), during the last 14 days and did not differ between the 2 subsamples (p=0.982) (Tab. III). Based on a generalized linear regression (GLR) model, lower education (t=2.54; p<0.012) and smoking (t=2.41; p<0.016) predicted the frequency of diet NA in the whole sample. The second question on the DDFQ asked hemodialysis participants to estimate their degree of NA to diet prescription during the last 14 days. The majority of the US sample reported significantly lower degrees of NA to diet compared with the German subsample (z=-4.61; p<0.0001) (Fig. 1). Education level was a significant predictor of diet NA in both groups but showed an opposing effect: in the US patients, higher education was correlated with a higher degree of diet NA (t=2.49; p<0.016); in the German sample, lower education predicted a higher degree of diet NA (t=2.29; p<0.022). Smoking (t=2.60; p<0.010) and diabetes (t=1.98; p<0.048) were also independent predictors of the degree of diet NA in the German sample, whereas for the US sample no other predictors were found. Comparison of non-adherence to fluid between the US and EU subsamples Median self-reported frequency of NA to fluid was assessed for the whole sample. It was 3 times, ranging be- 5

6 Kugler et al: Non-adherence in hemodialysis Fig. 2 - Comparison of degree of nonadherence (NA) to fluid between the US and German subsamples. tween 0 and 14 times (mean 4.93 ± 4.92), during the last 14 days and did not show significant differences between the 2 subsamples (p=0.401) (Tab. III). GLR model analysis revealed being single (t=2.68; p<0.008), male sex (t=2.64; p<0.009) and lower education (t=2.54; p<0.012) as predictor variables for frequency of NA to fluid in the whole sample. In a next step, hemodialysis patients were asked to describe their degree of NA to fluid during the last 14 days. The majority of the US sample reported significantly lower degrees of NA to fluid when compared with their German counterparts (z=-5.03; p<0.0001) (Fig. 2). A significant predictor for NA to fluid based on GLR models in the US sample was single marital status (t=2.36; p<0.022). TABLE III COMPARISON OF FREQUENCY OF NON-ADHERENCE TO DIET AND FLUID BETWEEN THE US AND GERMAN SUB- SAMPLES* Total sample US sample German sample Comparison (n=456) (n=113) (n=343) U p Value Self-reported frequency of non-adherence to diet during the last 14 days (%) No deviation times times times Self-reported frequency of non-adherence to fluid during the last 14 days (%) No deviation times times times *Total less than 100% refers to missing data. 6

7 JNEPHROL 0000; 00( 00) : In the German sample, male sex (t=2.59; p<0.010), lower education (t=2.64; p<0.009) and longer times on dialysis (t=2.01; p<0.046) predicted NA to fluid. Comparison of biochemical, biological and dialysis adequacy markers between the US and German subsamples Biochemical, biological and dialysis adequacy markers are displayed in Table IV. Biochemical markers for dietary NA included serum phosphorus and serum albumin. Both appeared to be significantly different between patient groups from the 2 countries (p<0.0001). The dialysis adequacy marker of Kt/V was higher in the US compared with the German subsample (p<0.0001). IDWG as a biological marker for fluid NA was higher in the US versus the German subsample (p<0.003). Comparison of self-reported non-adherence with biological and biochemical markers between the US and German subsamples Finally, self-reported NA with diet and fluid was compared between the 2 groups using biochemical, biological and dialysis adequacy markers. Although the US subsample had self-reported a lower NA with diet, phosphorus and albumin levels were higher, in contrast to the German subsample. The GLR model for phosphorus and albumin confirmed that US patients were less adherent with diet compared with their German counterparts (t=5.89; p<0.0001). Also in the US subsample, self-reported NA with fluid and IDWG as a marker for fluid NA contrasted: self-report was lower whereas IDWG was higher in comparison with the German subsample (t=4.66; p<0.0001). For both countries, Kt/V was not significantly different from self-reported diet NA. Di s c u s s i o n This is the first comparison study assessing centerbased, adult hemodialysis patients adherence to diet and fluid restrictions between 2 countries (United States and Germany) with distinct health care systems, in an attempt to look beyond patient-related factors, as suggested in the landmark report on adherence in chronic conditions by Sabate (1) on behalf of the WHO. Our findings indicate that executing lifestyle changes to adhere to diet and fluid restrictions accompanied with the treatment of ESRD by hemodialysis was difficult to maintain for a high percentage of patients in both countries the United States and Germany. Our data also suggest differences between the 2 patient groups including number of comorbidities, response rates and differences in diet and fluid NA. A significant finding in our study was that a large number of hemodialysis patients in both countries have difficulties maintaining their diet (80.4%) and fluid (75.3%) restrictions. These findings are in line with other studies using patients self-reports to determine prevalence of NA (15, 20). However, other research using laboratory values exclusively to estimate adherence in hemodialy- TABLE IV COMPARISON OF BIOLOGICAL AND BIOCHEMICAL MARKERS BETWEEN THE US AND GERMAN SUBSAMPLES Total sample US subsample German subsample U Comparison p Value Phosphorus (mg/dl) 4.70 ± ± ± < Albumin (mg/dl) 3.44 ± ± ± < Kt/V 1.27 ± ± ± < IDWG (kg) 2.29 ± ± ± <0.003 Values are means ± SD of the data, unless indicated otherwise. IDWG = interdialytic weight gain; SD = standard deviation. 7

8 Kugler et al: Non-adherence in hemodialysis sis patients (4, 6, 7) found lower prevalence of NA. This suggests that self-reports assessing NA in a neutral and nonthreatening way may have the potential to detect even minor deviations from the diet and fluid recommendations given to the individual. Laboratory data may be less sensitive as determinants of diet and fluid NA because they may be influenced by other variables such as phosphate binder adherence. Additionally, Kt/V may be influenced by the skill of the nephrologist in managing dialysis adequacy (12). A concerning finding of our study was the discrepancy between US versus German patients self-reported NA to diet and fluid when compared with the biochemical and biological markers. Participants from the United States may have been more inclined to provide a favorable response when asked about their adherence to diet and fluid restrictions. In the United States, dialysis patients do not routinely have 1 dialysis nurse who consistently provides nursing care during each dialysis session. Instead, patients are provided care by a team of technicians and nurses. Consequently, the therapeutic relationships in which adherence issues can be discussed in a nonthreatening manner are not routinely established in US dialysis centers. In the German patients, the key person to discuss critical health issues with is often the primary physician. In dialysis this person is often replaced by the center nephrologist. With dialysis lasting over years, therapeutic relationships are developed to discuss adherence issues in a nonjudgemental way. Another study, comparing NA in US versus European kidney transplant patients came to the same conclusion. The authors argue that transcultural factors, especially specific illness beliefs, might differ among patients from different countries (26). The cultural factor of illness beliefs falls into the WHO category of socioeconomic factors (1). Cultural factors may influence all of the WHO model categories and should be considered to be added to the WHO model. Culturally appropriate interventions can then be designed to improve adherence (27). Nutritional management appears to vary to some extent between the countries and may be related to cultural factors. The phosphorus and albumin content of the diet in both countries differs. Nutritional habits in the US subsample include eating more high-phosphorus-containing nutrients such as fast food products. On the other hand, cultural habits especially in this age group of the German sub-sample include more home-made nutrition containing fewer phosphorus supplements. However, this may change in the future with younger Germans developing busier lifestyles and using more industrially prepared products with higher phosphorus supplementation. Since nutritional status in hemodialysis has been shown to be strongly correlated with mortality rates in this population (4, 28-31), more research is needed with emphasis on the influence of the social environment. A recent systematic review (12) assessed the relationship between educational preparation and NA, with 21 of 23 studies documenting no statistically significant relationship. In our study, education appeared to predict NA to diet and fluid. However, our data showed contradictory findings for this variable when taking the country of origin for both cohorts into account. More specifically, higher education did predict NA to diet and fluid in the US subsample, whereas in German hemodialysis patients, lower education was found to indicate NA. This might be due to patients age in the German subsample and the associated societal and economic circumstances, since they grew up during or after World War II with limited access to higher education during this period of time. Limited evidence exists regarding the impact of the health care system on NA (26). However, because chronic hemodialysis is associated with high costs for insurance companies and for the chronically ill patient, the impact of this factor on NA seems to have been underestimated in the past. In the US subsample, about 99% of the participants have insurance coverage for the hemodialysis treatments and any medications administered during the treatment. The most common combination of insurance coverage is Medicare (which pays for 80% of the costs) and either Medicaid or another type of insurance (private or group insurance) which pays for the other 20% of costs. Insurance coverage for medications self-administered by the patient varies widely. Some patients may have a small co-pay of US $5-10, while others may pay close to US $1,000 for monthly medications such as phosphate binders (Norma Knowles, personal communication, December 22, 2009). Germany has a social welfare system with complete coverage of the costs for dialysis treatment with just 1 exception: there is a small amount of co-payment for medications for German chronically ill patients. Our study has several limitations. First of all, other aspects of adherence with the therapeutic regimen including medication adherence and appointment keeping may have influenced our findings, but were not controlled for in this study. More specifically, phosphorus level outside the reference values may have been influenced by phosphate binder medication NA. However, how much this influenced NA cannot be estimated from the data avail- 8

9 JNEPHROL 0000; 00( 00) : able. Secondly, skipped or shortened dialysis sessions may be in part responsible for higher IDWG levels measured in both subsamples, but were not controlled for in this study. Thirdly, nutritional habits are highly related to patients illness-specific knowledge and subsequent provision of patient education in the dialysis centers under study. Consequently, our findings may have been influenced by the presence and educational impact of a dietician or renal physician (32). Comorbidities were different in both subsamples with the US subsample having more illness demands. Finally, the samples may not be representative of all of the United States and Europe since the US subsample was from a single geographic area in the United States and the European sample was obtained from 2 geographic areas in Germany. However, this was an exploratory comparison study to identify if country-related differences were apparent. In conclusion, non-adherence persists as one of the most challenging tasks in the health care of patients with chronic conditions, including hemodialysis patients. Our findings suggest that patient and health care system related factors may contribute to non-adherence with diet and fluid restrictions. This study identified the need for a larger, fully powered study including diverse samples. Further development of interventional strategies should take transcultural issues into account. In addition, our findings indicate that adherence-improving interventions should be evaluated for their validity when applied in other patient populations in other countries. Institutional review board approval was obtained; this study was in adherence with the Declaration of Helsinki. Financial support: National Kidney Foundation (US data). None (German data). Conflict of interest statement: None declared. Address for correspondence: Christiane Kugler, PhD Division of Cardiac, Thoracic, Transplantation and Vascular Surgery Hannover Medical School Carl-Neuberg-Strasse 1 DE Hannover, Germany kugler.christiane@mh-hannover.de Re f e re n c e s 1. Sabate E. Adherence to long-term therapies: evidence for action. Geneva: World Health Organization; Kammerer J, Garry G, Hartigan M, Carter B, Erlich L. Adherence in patients on dialysis: strategies for success. Nephrol Nurs J. 2007;34: Walser M. Is there a role for protein restriction in the treatment of chronic renal failure? Blood Purif. 2000;18: Unruh ML, Evans IV, Fink NE, et al. Skipped treatments, markers of nutritional nonadherence, and survival among incident hemodialysis patients. Am J Kidney Dis. 2005;46: Tentori F, Hunt WC, Rohrscheib M, et al. Which targets in clinical practice guidelines are associated with improved survival in a large dialysis organization? J Am Soc Nephrol. 2007;18: Bame SI, Petersen N, Wray NP. Variation in hemodialysis patient compliance according to demographic characteristics. Soc Sci Med. 1993;37: Leggat JE, Orzol SM, Hulbert-Shearon TE, et al. Noncompliance in hemodialysis: predictors and survival analysis. Am J Kidney Dis. 1998;32: Ashurst IB, Dobbie H. A randomized controlled trial of an educational intervention to improve phosphate levels in hemodialysis patients. J Ren Nutr. 2003;13:

10 Kugler et al: Non-adherence in hemodialysis 9. Hecking E, Bragg-Gresham JL, Rayner HC, et al. Hemodialysis prescription, adherence and nutritional indicators in five European countries: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant. 2004;19: Sharp J, Wild MR, Gumley AI, Deighan CJ. A cognitive behavioural group approach to enhance adherence to hemodialysis fluid restrictions: a randomized controlled trial. Am J Kidney Dis. 2005;45: Denhaerynck K, Manhaeve D, Dobbels F, Garzoni D, Nolte C, De Geest S. Prevalence and consequences of nonadherence to hemodialysis regimens. Am J Crit Care. 2007;16: Russell CL, Knowles N, Peace L. Prevalence, predictors and correlates of non-adherence to a hemodialysis regimen: a review of the literature. J Nephrol Soc Work. 2007;27: Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353: Christensen AJ, Smith TW. Personality and patient adherence: correlates of the five- factor model in renal dialysis. J Behav Med. 1995;18: Vlaminck H, Maes B, Jacobs A, Reyntjens S, Evers G. The Dialysis and Fluid Non-adherence Questionnaire: validity testing of a self-report instrument for clinical practice. J Clin Nurs. 2001;10: Casey J, Johnson V, McClelland P. Impact of stepped verbal and written reinforcement of fluid balance advice within an outpatient haemodialysis unit: a pilot study. J Hum Nutr Diet. 2002;15: Durose CL, Holdsworth M, Watson V. Przygrodzka. Knowledge of dietary restrictions and the medical consequences of noncompliance by patients on hemodialysis are not predictive of dietary compliance. J Am Diet Assoc. 2004;104: Kugler C, Vlaminck H, Haverich A, Maes B. Nonadherence with diet and fluid restrictions among adults having hemodialysis. J Nurs Scholarsh. 2005;37(1): Barnett T, Yoong TL, Pinikahana J, Si-yen T. Fluid compliance among patients having haemodialysis: can an educational programme make a difference? JAN 2007; 61: Kara B, Caglar K, Kilic S. Nonadherence with diet and fluid restrictions and perceived social support in patients receiving hemodialysis. J Nurs Scholarsh. 2007;39: Russell CL, Whitlock R, Knowles N, Peace L, Tanner B, Hong TB. Rates and correlates of therapy non-adherence in adult hemodialysis patients. J Nephrol Soc Work. 2008;1: O Connor SM, Jardine AG, Millar K. The prediction of selfcare behaviours in end-stage renal disease patients using Leventhal s Self-Regulatory Model. J Psychosom Res. 2008;65: Lindberg M, Pruetz KG, Lindberg P, Wikström B. Interdialytic weight gain and ultrafiltration rate in hemodialysis: lessons about fluid adherence from a national registry of clinical practice. Hemodial Int. 2009;13(2): Polit DF, Beck CT. Nursing research: generating and assessing evidence for nursing practice. 8th ed. New York: JB Lippincott; National Kidney Foundation. Clinical practice guidelines for nutrition in chronic renal failure. Am J Kidney Dis. 2000;35:S17-S Denhaerynck K, Desmyttere A, Dobbels F, et al. Nonadherence with immunosuppressive drugs: US compared with European kidney transplant recipients. Prog Transplant. 2006;16: Russell CL. Culturally responsive interventions to enhance immunosuppressive medication adherence in older African American renal transplant recipients. Prog Transplant. 2006;16: Cohen SD, Kimmel PL. Nutritional status, psychological issues and survival in hemodialysis patients. Contrib Nephrol. 2007;155: Ford JC, Pope JF, Hunt AE, Gerald B. The effect of diet education on the laboratory values and knowledge of hemodialysis patients with hyperphosphatemia. J Ren Nutr. 2004;14: Leavey SF, Strawderman RL, Jones CA, Port FK, Held PJ. Simple nutritional indicators as independent predictors of mortality in hemodialysis patients. Am J Kidney Dis. 1998;31: Pifer TB, McCullough KP, Port FK, et al. Mortality risk and changes in nutritional indicators among hemodialysis patients in the DOPPS. Kidney Int. 2002;62: Kugler C, Saueressig U, Rausch, H, et al. Influence of patient education on nutrition management in hemodialysis patients. Nieren- und Hochdruckkrank-heiten. 2004; 33: Received: March 19, 2010 Revised: May 14, 2010 Accepted: June 16,