Original research; pancreatic enzyme replacement therapy: adherence to guidelines. (Insight).
Objective: To assess adherence to fat-based pancreatic enzyme doses in children with cystic fibrosis (CF).
Design, subjects and setting: Dietary and pancreatic enzyme intake and faecal fat balance studies were measured on a self-selected cohort of 38 children with CF. Each child acted as his or her own control. Data were collected through the hospital-based clinic approximately six months apart, before and after implementation of the Australian Pancreatic Enzyme Replacement Therapy Guidelines. The data analysed were of 29 subjects as complete data were not available on nine subjects.
Main outcome measures: The percentage of meals and snacks (occasions) which were accompanied by excessive and inadequate doses of pancreatic enzymes was calculated. Inadequate dosing was further classified as minor or major, depending on both the amount of lipase and percentage of occasions that were under-dosed. The lipid content of the wet stool was measured and expressed as percent faecal fat excretion.
Statistical analyses: Differences in pre- to post-intervention measurements were analysed using paired t-tests. The associations between percentage of faecal fat excretion and the percentage of occasions of inadequate and excessive dosing were assessed using correlation, linear and logistic regression analysis.
Results: The percentage of occasions accompanied by excessive or inadequate amounts of lipase decreased significantly pre- to post-intervention (P = 0.001 and P = 0.040, respectively, n = 29). However, the level of inadequate dosing after the intervention remained clinically unacceptable.
Conclusions: This study highlighted the importance of assessing and optimising adherence to doses on a meal and snack basis before efficacy of pancreatic enzyme replacement therapy guidelines can be established. (Nutr Diet 2002;59:260-4)
Key words: pancreatic enzyme replacement therapy, cystic fibrosis, adherence, dose
Eighty-five to 90% of individuals with cystic fibrosis (CF) have pancreatic insufficiency and require pancreatic enzyme replacement therapy (PERT) to manage maldigestion and malabsorption (1). Refinement of pancreatic enzyme preparations over the past two decades has resulted in intestinal absorption of dietary fat increasing from 80% to 90% of that ingested (2). The improvement in absorption has enabled individuals with CF-related pancreatic insufficiency to tolerate larger amounts of fat, and has also been associated with better nutritional status and survival rates (2, 3).
PERT requirements vary considerably, within and between individuals, due to a number of factors, including patient characteristics (degree of pancreatic insufficiency, the rate of gastric emptying, gastric and intestinal pH, small bowel motility, the amount and character of the intestinal mucus, bile excretion, formation of micelles with bile and time of ingestion of the capsules in relation to food) and the bioavailability of preparations (related to enzyme source, manufacturing process, stability of enzymes, enteric coating, formulation and dissolution characteristics and particle size). Doses of PERT have traditionally been based on body weight and, as pancreatic enzyme preparations were initially thought to be without adverse effects, the amount was increased until a reduction in symptoms associated with malabsorption (abdominal pain and loose stools) was achieved (4-6).
The incidence of fibrosing colonopathy, associated with high doses of pancreatic enzymes and/or the use of high strength preparations in the early 1990s (7-9), caused many centres treating CF to review the practices of their patients and collaborate to produce national guidelines (4,10,11). The major difference between the Australian PERT Guidelines for CF and those of other countries is that the dosage recommended for each meal and snack is dependent on the amount of dietary fat consumed, rather than on body weight. These fat-based doses are expressed as international units (IU) of lipase per gram of dietary fat, and more simply for individuals with CF as one capsule for a specific number of grams of fat, depending on the type of pancreatic enzyme preparation used and the individual's need (11,12). For example, a dose of 1000 IU of lipase per gram of fat is expressed as one capsule per 5 g of fat for preparations containing 5000 IU of lipase per capsule, and as one capsule per 10 g of fat for preparations co ntaining 10 000 IU of lipase per capsule.
Although research assessing the different methods of pancreatic enzyme dosing is scarce (13-15), the fat-based method is physiologically sound (4), and preferred (14), and warrants further investigation. A study was undertaken to assess the effect of the Australian PERT Guidelines' fat-based dosing on absorption. The purpose of this paper is to highlight factors related to adherence to fat-based pancreatic enzyme doses in a small group of children with CF.
The Australian PERT Guidelines for CF (11,12) were interpreted for individuals with CF and their carers in the 1998 version of the 'Go and Grow with CF' nutrition and behaviour intervention program (16). The up-dated program provides details regarding the distribution, administration and storage of pancreatic enzymes and includes resources to enable individuals with CF to match capsules to dietary fat intake.
All children aged two to 13 years attending the CF clinic at Princess Margaret Hospital, Perth, Australia, (n = 97) and/or their parents were given the opportunity to participate in the 'Go and Grow with CF' program between April and July 1998, in order to learn the Australian PERT Guidelines. Even though 'Go and Grow with CF' targets two- to 11-year-olds and their parents, children aged 12 and 13 years of age were also offered the 1998 revised version as the majority of these adolescents (10 out of 12) had participated in the clinical trial which evaluated the original program (16).
Information about how the change in PERT practices at the hospital would be assessed was provided to children and their parents at the same time that they received an invitation to participate in the home-based 'Go and Grow with CF' program.
Parents of a self-selected cohort of 38 children aged between two and 13 years, who were assessed as having pancreatic insufficiency when CF was diagnosed, agreed to a range of measurements being taken (e.g. height, weight, dietary and pancreatic enzyme intake and faecal output) approximately six months apart, before and after implementation of the Australian PERT Guidelines. Each child acted as his or her own control. Comparisons were made between pre- and post-intervention values.
Faccal fat balance studies
Parents were shown how to complete accurate dietary and pancreatic enzyme records of their child's intake for five days (e.g. by including details or packaging that described the type of food, indicating the cooking method and using household measures including scales where possible). Parents were advised to collect all stools from their child for four days, commencing after completion of the first 24 hours of dietary recording. The Food Works food composition database (Xyris Software, Brisbane, version 1.05 1997 ) was used by a dietitian to analyse the dietary records to determine each child's mean intake of fat. The lipid content of the wet stool was measured and expressed as percentage faecal fat excretion, i.e. the mean daily coefficient of fat absorption (17). The normal level is considered to be between 0 and 7% (18). Two different criteria, namely > 7% and > 10% faecal fat excretion, were used in the analysis as centres differ in what they consider as normal absorption for individuals with CF.
Traditionally, the practice at Princess Margaret Hospital was to recommend a weight-based PERT dose of approximately 5000 IU lipase per kg per day (i.e. one Pancrease (Janssen-Cilag, 5000 IU lipase per capsule) or half a Cotazym-S-Forte (Organon Australia, 10 000 IU lipase per capsule per kg per day). The dose was increased and adjuvant therapies (e.g. a [H.sub.2]-antagonist) considered if signs of possible malabsorption were evident. Individuals with CF and their parents were advised against self-adjusting the dose in the 1990s when fibrosing colonopathy was identified by other centres.
With the introduction of fat-based dosing, the CF clinic dietitian and gastroenterologist recommended a dose of either 1250 or 1700 IU lipase per g of dietary fat to each child, depending on the individual's pre-intervention dietary and pancreatic enzyme records and faecal fat excretion results. For example, if a dietary record indicated that a child was taking approximately 1700 IU lipase per g of dietary fat, and their faecal fat excretion was [less than or equal to] 7%, then they were advised to take a lower dose, such as 1250 IU lipase per g of fat. These doses (1250 and 1700 IU lipase per g of fat) corresponded to the more simple expression of one capsule for a specific number of grams of fat, depending on the type of pancreatic enzyme preparation used (i.e. a recommended dose of 1250 IU lipase per g of fat was achieved by taking one Pancrease capsule per 4 g of fat or one Cotazym-S-Forte capsule per 8 g of fat).
Pre- and post-intervention, the amount of lipase taken with each meal and snack was compared to the child's recommended fat-based dose. Such analyses have not been reported in the literature, so the investigators rated the dose of pancreatic enzymes taken as excessive if the subject took [greater than or equal to] 10 000 IU lipase per meal or snack more than his or her recommended dose, and as inadequate if he or she took [greater than or equal to] 5000 IU lipase per meal or snack below his or her recommended dose. The percentage of meals and snacks, or occasions, accompanied by excessive and inadequate doses of pancreatic enzymes was then calculated. The subject's level of inadequate dosing was further classified as minor or major, depending on both the amount of lipase and percentage of occasions that were under-dosed.
The children were weighed wearing underclothes on a digital scale (Wedderburn, NSW, Australia) to the nearest 0.1 kg. Stretched stature was measured in triplicate with a stadiometer (Holitan, Crymych, England) to the nearest 0.1 cm and the median value recorded. Weight and height z-scores were determined using the Ozgrow software program (Ozgrow, Camperdown, NSW, 1995) which compared the subjects' measurements to the National Centre for Health Statistics (NCHS) age- and gender-norms (19).
Factors which could interfere with the effects of PERT (adjuvant therapies to improve the pH of the gastrointestinal tract, gut surgery and tube feeding) were recorded pre- and post-intervention.
Parents of 39 children with PI who completed the 'Go and Grow with CF' program, regardless of whether or not they conducted faecal fat balance studies, were interviewed by telephone. Open-ended questions were used to determine parents' perceptions about matching pancreatic enzyme capsules to grams of dietary fat.
Informed consent and ethics approval were not required by the hospital administration as this study was primarily an assessment of a change in clinical practice. After providing feedback to parents about their child's results, patient data were collated and analysed anonymously.
The Statistical Package for Social Sciences (SPSS Inc, Chicago, version 6.1.3 1995) was used to analyse differences in pre- to post-intervention measurements by paired t-tests. The associations between percentage of faecal fat excretion and explanatory variables (e.g. amount of lipase, percentage of occasions of inadequate dosing and adherence) were assessed using correlation, linear and logistic regression analysis.
Complete data were not available for nine subjects because pre-intervention dietary records were incomplete (n = 1) or post-intervention data were not collected when children transferred interstate during the intervention period (n = 4) and when parents chose not to complete the second faecal fat balance study (n = 4). Thus the data analysed were of 29 subjects (14 males, 15 females) with a mean age of 7.6 years (range 1.6 to 13.6 years).
Subjects were maintained on the same pancreatic enzyme preparation throughout the study. Nineteen subjects took Pancrease (10 males, 9 females) and 10 subjects used Cotazym-S-Forte (4 males, 6 females). The mean level of faecal fat excretion of subjects on Pancrease was not significantly different to those on Cotazym-S-Forte.
Four subjects had a short-gut and one of these was on long-term nocturnal gastrostomy feeds. During the study, five subjects were taking either an [H.sub.2]-antagonist (ranitidine), synthetic prostaglandin (misoprostil) or a proton-pump inhibitor (omeprazole).
Pre-intervention, the dose recommended to 22 subjects was 1250 IU lipase per g of dietary fat (Pancrease n = 17, Cotazym-S-Forte n = 5). The remaining seven subjects were prescribed a dose of 1700 IU lipase per g of dietary fat (Pancrease n = 2, Cotazym-S-Forte n = 5).
Pre- and post-intervention, most of the parents chose to perform the faecal fat balance studies during the school holidays or over a long weekend for ease of faecal collection at home.
Paired t-tests indicated that the mean increase in fat intake, decrease in lipase intake and improvements in percentage of faecal fat excretion, z-height and z-weight scores were non-significant (P > 0.05) (Table 1). Comparisons between the individual's recommended dose and their actual dose by paired t-tests indicated that the percentage of occasions accompanied by excessive or inadequate amounts of lipase decreased significantly pre- to post-intervention (P = 0.001 and P = 0.040, respectively) (Table 1). Although the percentage of occasions of inadequate dosing improved, the level after the intervention was clinically unacceptable (21%, representing one meal or snack every day). Excessive and inadequate dosing occurred at both meals and snacks.
Correlation, linear and logistic regression analysis indicated that percentage of faecal fat excretion was not significantly associated with either the percentage of meals and snacks which were inadequately dosed, the total amount of lipase that was inadequate or the level of inadequate dosing. Both pre- and post-intervention, subjects classified as having a major level of inadequate dosing (based on the amount of lipase and percentage of occasions of inadequate dosing) were more likely to have abnormal faecal fat excretion (> 7%) than those with minor levels of inadequate dosing, however, this was not statistically significant.
There was a strong preference for the new method of pancreatic enzyme dosing among parents who completed the 'Go and Grow with CF' program with their child. Thirty-seven of the 39 parents interviewed indicated that they prefer doses based on the fat content of each meal and snack over the previous method (a daily body weight-based dose) because of:
* a perceived improvement in symptoms associated with malabsorption in their child, such as less abdominal pain, less oily or loose bowel actions, and less constipation;
* their increased confidence with enzyme therapy; and
* their ability to decrease the number of capsules at some meals and snacks without any adverse effects.
The two parents who did not prefer the new method indicated that they were likely to, once they became more familiar with the fat content of food and drinks.
The levels of inadequate and excessive dosing in this study improved pre- to post-intervention, but the level of inadequate dosing remained clinically unacceptable. The 21% of occasions which were under-dosed represented one meal or snack per day, on average. It is likely that inadequate dosing in the general CF population is even more clinically significant, given that parents who agreed to participate in this study may have been motivated to achieve optimum pancreatic enzyme dosing. The continuing high frequency of inadequate dosing, together with the amount of lipase that was deficient, may have limited this study in its ability to indicate the effect of fat-based dosing on fat absorption. The small sample size, and pre-intervention mean level of faecal fat excretion being near normal (0% to 7%), also would have contributed to the lack of significant improvements in the outcomes measured.
This study highlights the need to determine and address reasons for non-adherence, before the effects of the various methods of PERT dosing on fat absorption can be assessed. Excessive and inadequate pancreatic enzyme dosing at particular meals and snacks may be due to factors such as poor knowledge about the fat content of food and fluids (20-23), parental hesitation to redistribute enzymes according to fat content and inadequate supervision of children taking capsules.
Knowledge deficits became evident when the dietitian provided feedback to parents about the appropriateness of the enzyme doses recorded during the faecal fat balance studies. Post-intervention, many parents indicated that they did not realise exactly how high the fat content was of some very high fat foods, such as chocolate, nuts, doughnuts, '2 minute noodles', takeaway meals, chips and chicken nuggets. This continuing lack of knowledge was not expected as several strategies had been implemented to enhance children's and parents' awareness of the fat content of food and fluids. Parents had received individually tailored feedback about matching fat intake more appropriately (based on the dietary records completed prior to the intervention) and families had been instructed about how to quantify the amount of fat in food, using resources such as the 'Go and Grow with CF' program material, fat counter booklets and labels on food packaging. It is possible that non-adherence to fat-based doses was a consequence o f parents either not referring to the resources or because they were hesitant to change the amount of pancreatic enzymes taken, by redistributing capsules throughout the day or by varying the dose within and between meals and snacks, without more direct guidance from a health care provider.
Parents' hesitations about changing PERT practices became obvious when a six-year-old boy took the upper limit of his usual snack dose when he ate a large amount (3/4 cup) of roasted and chocolate-coated nuts on two occasions during the post-faecal fat balance study. The subject's post-intervention faecal fat excretion level was excessive (20%), which was not surprising given that fat-based dosing indicated that he may have required three times his usual amount of lipase for this type of snack. This example illustrates the importance of objective assessments of nutrition and PERT knowledge until an adequate level is achieved and is successfully translated into practice. Regular education and support of families may assist the process of change by increasing levels of confidence and facilitating the development of necessary skills.
Inadequate supervision during the post-intervention faecal fat balance studies was a third reason for inadequate dosing. Unforeseen circumstances, which frequently occur in the clinical setting (e.g. holiday outings, parties and parents being required to work unexpectedly), resulted in some children self-managing their dietary intake and fat-based dosing during the faecal fat balance study. If children had not yet mastered fat-based dosing then they may have taken their previous usual dose for meals and snacks when they were not being supervised. This may have been inadequate for the amount of fat consumed.
Given the high level of non-adherence in this study and the numerous daily demands placed on children and their parents to conduct CF-related therapies (chest treatments, medication and diet), fat-based dosing may be considered as requiring too much effort to be warranted. However, the need for further investigation into the effects of fat-based dosing on absorption is supported foremost by the strong preference for the new method among parents, and secondly by the high prevalence of above-normal faecal fat excretion levels despite advances in the treatment of CF, including the use of adjuvant therapies (approximately three-quarters of the subjects had excretion levels > 7%). The time taken to determine the amount of pancreatic enzymes required is likely to decrease considerably once families are familiar with the fat content of the child's usual meals and snacks (which are often similar from one day to the next). Another factor that supports the implementation of fat-based dosing is that improved levels of k nowledge about the composition of food may assist children in achieving a high energy, high fat diet as recommended for individuals with CF.
It will be important for future assessments of PERT methods to minimise the confounding effect of poor adherence to doses by including assessments of dose on a meal and snack basis, rather than just on daily intake (14,15). With regard to implementation of fat-based dosing, we recommend that individuals with CF or their parents complete a series of detailed three- to five-day dietary and pancreatic enzyme records (forms available from the authors) until a high level of adherence is achieved, such as no more than one meal or snack per day being under-dosed by 5000 IU of lipase or less. This process of self-monitoring may become an important factor in assisting individuals with CF in changing their pancreatic enzyme dosing practices. The dietary and pancreatic enzyme records will also enable the dietitian to provide periodic, individualised, objective feedback about the appropriateness of pancreatic enzyme doses. Studies based on these recommendations are needed to establish evidence-based PERT dosing practices and to ascertain if fat absorption can be improved to near normal levels (0 to 7% faecal fat excretion) in the majority of individuals with CF.
Table 1 Pre- to post-intervention changes in fat and lipase intakes, absorption and anthropometric characteristics of children with cystic fibrosis-related pancreatic insufficiency, aged one to 13 years (n = 29) (a) Measure Pre-intervention Post-intervention Fat intake (g/day) 91 [+ or -] 27 94 [+ or -] 32 Lipase intake (IU/kg/day) (c) 6390 [+ or -] 2751 6202 [+ or -] 2019 (IU/g dietary fat) 1532 [+ or -] 474 1583 [+ or -] 308 Dose excessive (d) (% meals and 25 [+ or -] 23 9 [+ or -] 12 snacks/day) Dose inadequate (e) (% meals and 30 [+ or -] 17 21 [+ or -] 13 snacks/day) FFE (f) (co-efficient %) 11.0 [+ or -] 6.5 10.3 [+ or -] 4.2 Height (z-score) -0.5 [+ or -] 1.0 -0.4 [+ or -] 1.1 Weight (z-score) -0.5 [+ or -] 0.9 -0.4 [+ or -] 0.9 Measure 95% CI (b) P-value Fat intake (g/day) -9, 4 0.39 Lipase intake (IU/kg/day) (c) -469, 846 0.56 (IU/g dietary fat) -233, 130 0.56 Dose excessive (d) (% meals and 7, 24 0.001 snacks/day) Dose inadequate (e) (% meals and 0.4, 17 0.040 snacks/day) FFE (f) (co-efficient %) -1.4, 2.9 0.500 Height (z-score) -0.1, 0.1 0.096 Weight (z-score) -0.2, 0.1 0.250 (a)Mean [+ or -] SD (b)95% confidence interval for the difference between pre- and post- intervention mean values. (c)International units (d)[greater than or equal to]10 000 IU lipase in excess (e)[greater than or equal to]5 000 IU lipase inadequate (f)Faecal fat excretion, percentage of dietary fat excreted.
The authors are very grateful to Barbara Keating, dietitian and research assistant, for dietary analysis; to staff of the Department of Clinical Chemistry, Princess Margaret Hospital, for conducting the faecal fat balance studies; to all the children and their parents who participated in the study; and to Solvay Pharmaceuticals and Organon (Australia) for financial support.
(1.) Fitzsimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993;122:1-9.
(2.) George DE, Mangos JA. Nutritional management and pancreatic enzyme therapy in cystic fibrosis patients: state of the art in 1987 and projections into the future. J Pediatr Gastroenterol Nutr 1988;7:49S-57S.
(3.) Corey M, McLaughlin FJ, Williams M, Levison H. A comparison of survival, growth and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol 1988;41:583-91.
(4.) Borowitz DS, Grand RJ, Dune PR. Use of pancreatic enzyme supplements for patients with cystic fibrosis in the context of fibrosing colonopathy. J Pediatr 1995;127:681-4.
(5.) Brady MS, Rickard K, Yu P, Eigen H. Effectiveness of enteric coated pancreatic enzymes given before meals in reducing steatorrhea in children with cystic fibrosis. J Am Diet Assoc 1992;92:813-7.
(6.) Taylor CJ, Hillel PG Ghosal S, Frier M, Senior S, Tindale WB, et al. Gastric emptying and intestinal transit of pancreatic enzyme supplements in cystic fibrosis. Arch Dis Child 1999;80:149-52.
(7.) Campbell CA, Forrest J, Musgrove C. High-strength pancreatic enzyme supplements and large-bowel stricture in cystic fibrosis. Lancet 1994;343:109-10.
(8.) Oades PJ, Bush A, Ong PS, Brereton RJ. High-strength pancreatic enzyme supplements and large-bowel stricture in cystic fibrosis. Lancet 1994;343:109.
(9.) Smyth RL, van Velzen D, Smyth AR, Lloyd DA, Heaf DP. Strictures of ascending colon in cystic fibrosis and high-strength pancreatic enzymes. Lancet 1994;343:85-6.
(10.) Committee on the Safety of Medicines. Report of the pancreatic enzymes working party. London: Medicines Control Agency; 1995.
(11.) Anthony H, Collins CE, Davidson G, Mews C, Robinson P, Shepherd D, et al. Pancreatic enzyme replacement therapy in cystic fibrosis: Australian guidelines. J Paediatr Child Health 1999;35:125-9.
(12.) Stapleton D, Anthony H, Collins C, Powell E, King S, Mews C. Clinical practice guidelines: Implementing the Australian pancreatic enzyme replacement therapy guidelines for cystic fibrosis. Aust J Nutr Diet 1999;56:91-6.
(13.) Constantini D, Padoan R, Curcio L, Giunta A. The management of enzymatic therapy in cystic fibrosis patients by an individualized approach. J Pediatr Gastroenterol Nutr 1988;7:36S-9S.
(14.) Durie P. Kalnins D, Ellis L. Uses and abuses of enzyme therapy in cystic fibrosis. J R Soc Med 1998;91:Suppl:2-13.
(15.) Kalnins D, Dune PR, Ellis L. Pancreatic enzymes-evaluation of current practice in a large CF clinic. Pediatr Pulmonol 1995;20:266.
(16.) Stapleton D, Tunnecliffe L, McGuiness D, Sheriff J, Sly P. Development of a nutrition and behaviour intervention program: Go and Grow with CF. Aust J Nutr Diet 1998;55:130-7.
(17.) van de Kamer JK, ten Bokkel Huinink H, Weyers HA. Rapid method for the determination of fat in feces. J Biol Chem 1949;177:347-55.
(18.) Couper R, Dune PR. Pancreatic function tests. In: Walker WA, Durie PR, Hamilton JR, Walker-Smith JA, Watkins JB. editors. Pediatric Gastrointestinal Disease. Philadelphia: BC Decker Inc; 1991.
(19.) National Center for Health Statistics. Growth curves for children birth-18 years. Washington: National Center for Health Statistics; 1977.
(20.) Basketter H, Riordan A, Bilton D. Knowledge of pancreatic enzyme supplementation in adult cystic fibrosis patients. 21st European CF Conference; Davos, Switzerland; 1997. Abstract p.123.
(21.) McCabe H. Cystic fibrosis--what nutritional knowledge do patients have? J Hum Nutr Diet 1996;9:479-86.
(22.) Stapleton DR, Gurrin LC, Zubrick SR, Silburn SR, Sheriff JL, Sly PD. What do children with cystic fibrosis and their parents know about nutrition and pancreatic enzymes? J Am Diet Assoc 2000;100:1494-500.
(23.) Yuill R, Innes JA, Greening AP. Assessment of adult cystic fibrosis patients' knowledge and understanding of diet and pancreatic enzyme administration. 21st European CF Conference; Davos, Switzerland; 1997. Abstract p.170.
Princess Margaret Hospital for Children, Perth, Western Australia
D. Stapleton, PhD, Dietitian
C. Mews, FRACP, Gastroenterologist
Princess Margaret Hospital for Children and School of Population Health, University of Western Australia, Perth
M. Bulsara, MSc, Biostatistician
School of Public Health, Curtin University of Technology, Perth
J. Sherriff, PhD, Senior Lecturer
Division of Clinical Sciences, TVWT Institute for Child Health Research and Centre for Child Health Research, University of Western Australia, Perth
Peter Sly, MD, Division Head
Correspondence: D. Stapleton, Department of Nutrition and Dietetics, King Edward Memorial and Princess Margaret Hospitals, PO Box D184, Perth WA 6840.
|Printer friendly Cite/link Email Feedback|
|Publication:||Nutrition & Dietetics: The Journal of the Dietitians Association of Australia|
|Date:||Dec 1, 2002|
|Previous Article:||Constipation and diet in a community sample of older Australians. (Original Research).|
|Next Article:||Letter to the Editor.|