Adherence to Nutrition Supplements Among Patients With a Fall-Related Lower Limb Fracture
* Department of Rehabilitation and Aged Care and Correspondence: Maria Crotty, FAFRM(RACP), Flinders University Department of Rehabilitation and Aged Care, C-Block, Repatriation General Hospital, Daws Road, Daw Park, South Australia 5041. Electronic mail may be sent to maria.crotty{at}flinders.edu.au. Background: The purpose of this study was to provide a detailed evaluation of adherence to nutrition supplements by patients with a lower limb fracture. Methods: These descriptive data are from 49 nutritionally "at-risk" patients aged 70+ years admitted to the hospital after a fall-related lower limb fracture and allocated to receive supplementation as part of a randomized, controlled trial. Supplementation commenced on day 7 and continued for 42 days. Prescribed volumes aimed to meet 45% of individually estimated theoretical energy requirements to meet the shortfall between literature estimates of energy intake and requirements. The supplement was administered by nursing staff on medication rounds in the acute or residential care settings and supervised through thrice-weekly home visits postdischarge. Results: Median daily percent of the prescribed volume of nutrition supplement consumed averaged over the 42 days was 67% (interquartile range [IQR], 31–89, n = 49). There was no difference in adherence for gender, accommodation, cognition, or whether the supplement was self-administered or supervised. Twenty-three participants took some supplement every day, and a further 12 missed <5 days. For these 35 "nonrefusers," adherence was 82% (IQR, 65–93), and they lost on average 0.7% (SD, 4.0%) of baseline weight over the 6 weeks of supplementation compared with a loss of 5.5% (SD, 5.4%) in the "refusers" (n = 14, 29%), p = .003. Conclusions: We achieved better volume and energy consumption than previous studies of hip fracture patients but still failed to meet target supplement volumes prescribed to meet 45% of theoretical energy requirements. Clinicians should consider alternative methods of feeding such as a nasogastric tube, particularly in those patients where adherence to oral nutrition supplements is poor and dietary intake alone is insufficient to meet estimated energy requirements.
Osteoporotic lower limb fracture is a significant cause of morbidity and
mortality in older people, with up to one-third of patients dying in the first
year postfracture and many failing to regain their mobility or return to
independent living.1
In addition, the costs of hospitalization, postfracture care, and
rehabilitation in the community are
high.1
Undernutrition is evident in 38%–57% of elderly (age If oral nutrition supplementation is to improve health outcomes, the first prerequisite is that the supplement prescription must provide enough nutrients, particularly energy and protein, for a sufficient length of time to at least meet or exceed requirements, depending on whether maintenance or nutrition rehabilitation, respectively, is required. Second, adherence to the nutrition prescription to ensure adequate intake of supplement is critical. A recent Cochrane5 review of randomized, controlled trials of oral protein-energy supplements in elderly hip fracture patients highlights a range of study quality limitations that contribute to the conclusion that evidence of the effectiveness of nutrition supplementation after a fall-related fracture in improving health outcomes remains very weak.5 Individual studies contributing to the review have reported some beneficial effects of oral nutrition supplementation6,7 including reduced length of stay and complications such as delirium, urinary tract infection, and bed sores. Most recently, Bruce et al8 in 2003 reported an intention-to-treat analysis of a quasirandomized controlled trial in older hip fracture patients receiving 235 mL nutrition supplement per day (1478 kJ; 352 kcal) for 28 days. The basis for this prescription was not provided. No effect on function, length of hospital stay, or mortality was observed. The authors contributed the lack of effect, at least in part, to poor adherence (73%) to the oral supplement. However, dietary intake was not reported, and hence overall achievement of total energy requirements is unclear and may also have contributed to the lack of effect. In contrast, Bastow et al9 were able to show improvements in anthropometric indicators, rehabilitation time, and length of hospital stay in malnourished older patients post–hip fracture who received 1 L of nutrition supplement (4200 kJ; 1000 kcal) overnight via nasogastric tube for a median of 28 days, with usual inpatient diet offered during the day. It is likely that overcoming the adherence issues associated with oral supplements by use of nasogastric feeding to achieve a fourfold higher level of nutrition supplementation is a key factor in the observation of clear effects reported in this study. However, although Bastow et al9 reported minimal side effects, 22% of patients were unable to tolerate the feeding protocol for unspecified reasons. Adherence with nutrition supplement prescriptions is notoriously poor, and thus strategies to improve, monitor, and report adherence are important clinical and study-design issues.5,10 The Cochrane review5 identifies poor adherence, limited adherence data reporting, and failure to provide intention-to-treat analyses as major issues in supplementation trials. In the few studies that have reported directly on adherence,6–8,10,11 definitions and monitoring strategies are variable, and thus there are limited data to inform clinicians and researchers. The aim of this paper is to provide a detailed descriptive evaluation of adherence to oral nutrition supplements in a sample of older patients after lower limb fracture participating in a randomized, controlled trial.
Participants The descriptive data reported in this paper are from participants in the ENERGII Trial (Evaluation of Nutrition and Exercise as Geriatric Injury Interventions) involving 100 older adults admitted to Flinders Medical Centre (FMC) in Bedford Park, South Australia, after a fall-related lower limb fracture between September 2000 and October 2002. This randomized, controlled trial evaluated an individualized, 6-week, oral supplementation program or a progressive resistance lower limb strength-training program. Eligibility criteria included 70 years or older and at risk of malnutrition, defined by a midarm circumference (MAC) less than the 25th percentile of a large representative sample of older Australians (27.0 cm and 26.3 cm for men and women, respectively).12 MAC was used as a practical alternative to body mass index (BMI) for eligibility screening due to practical difficulties in obtaining accurate weight and height measurements without discomfort in older patients immediately after lower limb fracture with or without surgery. The World Health Organization recommends the use of MAC for screening older adults for nutrition interventions,13 and Bastow et al9 reported MAC was predictive of poorer intakes and health outcomes after hip fracture. Serum proteins such as albumin were not used for screening due to potential confounding of the acute phase response to injury and surgery.14 Patients were excluded from participating in the study if they were not medically stable within 7 days after admission or had cancer, chronic renal failure, unstable angina, or unstable diabetes. This paper reports on the supplement intake of 49 participants (11 men, 38 women) who were randomized to receive the 6-week oral supplementation alone (n = 25) or in combination with the exercise program (n = 24). Ethical approval was obtained from Flinders Clinical Research Ethics Committee. Written informed consent was obtained from each participant or next of kin.
Design and Delivery of the Oral Nutrition Supplement Intervention
Estimates of average daily energy intake in similar patients from diet alone over a variable number of days during the first 2 weeks post–lower limb fracture range from 3230 kJ (770 kcal)16 to 4620 kJ (1100 kcal).6 This range represents approximately 55%–65% of energy requirements estimated as above (using weight range of 50–65 kg). Unpublished pilot data using a weighed plate–waste food record from patients 70+ years during the first week post–hip fracture at FMC (n = 11) confirmed an average intake of 45%–62% of similarly estimated energy requirements. In summary, supplement volumes prescribed in this study aimed to meet 45% of individually estimated total energy requirements and ranged 580 mL–800 mL per day of the 6.3 kJ (1.5 kcal)/mL supplement. Supplement administration commenced day 7 postinjury, which for some patients was also postdischarge from acute care. Day 7 was chosen to coincide with commencement of resistance training and to provide adequate time for informed consent, particularly from caregivers of cognitively impaired participants, thus ensuring a consistent intervention schedule across all groups. To promote tolerance and adherence, a number of practical and clinically feasible strategies were used. A choice of 6 flavors of the nutrition supplement was offered to participants by the study dietitian in an effort to reduce flavor fatigue. The study dietitian ordered on a supplement prescription chart 25% of total prescribed daily volume to be dispensed by nursing staff 4 times per day from the medication cart on rounds ("supervised administration"). Our pilot observational comparison study demonstrated that this strategy compared with usual practice of supplement delivery with meals almost doubled the percentage of prescribed supplement taken over 3 days (79% vs 44%, n = 38, p = .01).17 Upon discharge to home, the supplement was scheduled to be given twice a day or more (total volume prescribed remained constant) in consultation with the participant or caregiver ("self administration"). For those discharged to residential care, administration of the supplement remained via the medication cart ("supervised administration"). Postdischarge, all participants received thrice weekly home visits to match participant contact of the exercise intervention and hence reduce any attention effect, encourage supplement consumption, change flavors if required, and monitor adherence.
Demographic Data, Cognition, Body Mass Index, Serum Prealbumin Level, Supplement, and Voluntary Energy and Protein Intake The volume of supplement prescribed and consumed was documented 4 times daily by nursing staff and confirmed by the research staff using counts of supplement packages (used and unused). During the thrice weekly home visits, research staff recorded remaining supplement packages to confirm the daily volume of supplement taken as recorded by the participant or caregiver in the study manual. Overall adherence is reported as the median daily supplement consumption calculated over the total 42 days of intervention, expressed as a percent of daily volume prescribed. The median was calculated over 27 days for 1 participant who died early on day 28. Similarly, median daily percent supplement consumed over each week of the intervention was calculated. The number of days on which no supplement was taken was recorded as was the last day on which supplement was taken, if before completion of the intervention. "Refusers" were defined as those who refused supplement for 5 days or more of the 42-day intervention, and conversely, "nonrefusers" refused supplements for only 0–4 days of the intervention period. Estimated voluntary daily inpatient energy and protein intake was available for 41 out of 49 participants receiving supplement commencing upon consent, between day 2 and day 6 postinjury, and continuing for up to 5 consecutive days. FMC provides a menu based on a schedule of 3 meals per day to all inpatients, and no specific strategies were used to promote volitional dietary intake. Intake was assessed using plate waste methodology and snack record charts and for practical reasons was discontinued on discharge from the acute care setting. Average daily energy and protein intake was calculated using dietary analysis software: SERVE Nutrition Management System, Version 3.0, M & H Williams Pty Ltd, Sydney, Australia.
Statistical Analyses
Table 1 outlines the baseline characteristics of 49 participants who received the nutrition intervention. There were no differences between men and women for age, admission accommodation, cognition, BMI, serum prealbumin levels, or MAC.
Table 2 provides details of supplement prescription and consumption. Participants were prescribed between 580 and 800 mL per day of the supplement according to individual estimated energy requirements, equating to at least 3654 kJ (870 kcal) of energy per day. Overall, the median daily proportion of the prescribed volume of supplement consumed calculated over the 42-day intervention was 67% (IQR 31–89). Figure 1 shows that 30 (61%) participants were able to achieve levels of adherence >60% of prescribed volume, with only 2 participants unwilling to take any supplement at all. There was no difference in the percent of prescribed supplement consumed by gender (Mann-Whitney Z = –1.270, p = .204), admission accommodation (Mann-Whitney Z = –0.875, p = .382) or cognition (Mann-Whitney Z = –0.616, p = .538). Those participants aged 70–84 years were able to consume 78% (IQR 59–92) of prescribed volume compared with 45% (IQR 19–82) in those aged 85 years and above (Mann-Whitney Z =–2.02, p = .043). Median daily supplement consumption was similar for supervised vs self-administration, 74% (IQR 33–86) and 75% (IQR 32–97), respectively, Wilcoxon Z = –0.422, p = .673. There was no difference across the weeks in median daily percent of supplement consumed calculated over each week for all participants (p = .098; n = 49) or for "nonrefusers" only (p = .813; n = 35; Figure 2).
The median number of days supplement was taken was 41 (IQR 31–42),
and the frequency of the total number of days that supplement was consumed is
shown in Figure 3, with 35
participants consuming supplement on 35 days or more. Twenty-six (53%)
participants failed to consume any supplement on at least 1 day. There were 14
(29%) "refusers" who failed to take supplement on 5 or more days
and 35 (71%) "nonrefusers." The baseline characteristics of the
"refusers" and "nonrefusers" are shown in
Table 3. The
"refusers" were significantly older (p = .045). The
median daily percent of the prescribed volume of supplement consumed computed
over the 42-day intervention by the "refusers" was 16% (IQR
2–40, n = 14). Four "refusers" consistently declined to take
supplement from days 1–3 of the intervention, 1 "refuser"
failed to consume supplement between days 4 and 6 and again between days 8 and
13, whereas the remaining 9 "refusers" ceased taking the
supplements completely some time after day 16 of the intervention. Of these 9
"refusers," 1 died at day 28 (had previously refused
The median daily percent of the prescribed volume of supplement consumed computed over the 42-day intervention by the "nonrefusers" was 82% (IQR 65–93, n = 35). Twenty-three "nonrefusers" took supplement every day (percent prescribed volume of supplement taken daily, 89%, IQR 75–97) and 12 missed 1–4 days of supplement (percent prescribed volume of supplement taken daily, 66%, IQR 50–78). Twenty-nine (83%) "nonrefusers" compared with 6 (43%) "refusers" (p = .012) were discharged from hospital to the community where the nutrition supplement was self-administered.
Average daily dietary energy and protein intake over a median of 5 days (IQR 3–5) per participant was available for 41 out of the 49 participants (84%) before discharge from the acute care setting. Twenty of these participants had a median of 1 day (IQR 0–2.5) of supplement provided during the dietary intake assessment. Overall voluntary daily energy intake (excluding supplement) was 3597 kJ; 856 kcal (IQR 3054; 727–4231; 1007) or 3393 kJ; 808 kcal (IQR 2314; 551–4254; 1013) for "refusers" and 3820 kJ; 910 kcal (IQR 3188; 759–4237; 1009) for "nonrefusers," p = .218. Overall voluntary daily protein intake (excluding supplement) was 37 g (IQR 25, 47) or 32 g (IQR 22–47) for "refusers" and 37g (IQR 29–48) for "nonrefusers," p = .353. Voluntary energy intake (excluding supplements) as percent estimated energy requirements according to the Schofield15 equation without adjustment was 69% (IQR 47–91) for the "refusers" compared with 75% (IQR 61–87) for the "nonrefusers." "Refusers" and "nonrefusers" achieved 37% (IQR 25–49) and 41% (IQR 33–47), respectively, of energy requirements when adjustments were included for activity level, trauma, and weight gain, p = .466. Voluntary protein intake (excluding supplements) as percent estimated protein requirements (1 g/kg/day) was 70% (IQR 44–98) for the "refusers" compared with 64% (IQR 50–85) for the "nonrefusers."
Of participants who consumed supplements on days of dietary intake
assessment (n = 20), 14 participants were able to achieve
At the end of the 6 weeks of supplementation (7 weeks postfracture) the "refusers" (n = 11: 2 withdrawn, 1 died) had lost 5.5% (SD 5.4%) of baseline weight, whereas the "nonrefusers" (n = 35) had lost 0.7% (SD 4.0%), p = .003.
This study provides the most detailed adherence data to date for a randomized, controlled trial of oral nutrition supplementation in elderly, nutritionally at-risk patients after a lower limb fracture. To our knowledge, it is the first study to attempt to take into account variability in individual energy requirements in the supplement prescription rather than a `1 dose fits all' approach and hence evaluate adherence relative to nutrition requirements in addition to volume prescribed. Only 1 other study8 includes cognitively impaired patients. Overall adherence to oral supplements was two-thirds of the volume prescribed, and as a result, these patients are at risk of failing to meet the estimated shortfall between individual energy requirements and dietary intake. Even patients who rarely (4 days or less) refused to take supplement had a median adherence rate of 82% of volume prescribed. Four patients (8%) refused to take supplements within the first 3 days, and a further 10 stopped taking supplements completely some time after 2 weeks, representing an overall refusal rate of close to 1 in 3. Only 5 published papers report adherence data in hip fracture patients receiving oral nutrition supplements.6–8,10,11 All of these papers define adherence simply in terms of number of cans or doses taken relative to number prescribed, rather than relative achievement of volume or energy prescription. Inclusion or otherwise of partially consumed cans is not explicit. Delmi et al6 reports 100% adherence in 27 hip fracture patients prescribed 250 mL (1 can, 4.2 kJ; 1 kcal/mL) of supplement, given at 8 PM over a mean of 32 days in hospital. This compares with a median prescribed volume of 640 mL (6.3 kJ; 1.5 kcal/mL) over 42 days across both the hospital and community setting in our study. The participants in the Delmi et al6 study were of similar age to those in the current study, but they were not selected to be nutritionally at-risk and none were cognitively impaired. Despite our comparatively lower level of adherence (67% vs 100%), we were able to achieve greater net volume intake (454 mL vs 250 mL) of a more energy-dense oral supplement (6.3 kJ; 1.5 kcal/mL vs 4.2 kJ; 1 kcal/mL) and hence almost 3 times greater intake of energy (2862 kJ; 681 kcal vs 1050 kJ; 250 kcal) from oral supplements. Williams et al10 prescribed 500–750 mL (2–3 cans, 4.2 kJ; 1 kcal/mL) of supplement per day between meals for 49 nutritionally at-risk, cognitively intact women over 60 years hospitalized for a mean of 21 days after a fracture. The reported early refusal (minimal total intake <100 mL in first 2 days and subsequent ongoing refusal) rate of 22% was slightly lower than 29% in our study (14 participants took supplement for <5 days) and may reflect a greater degree of supervision for inpatients only and the shorter time frame. Espaulella et al7 reported that 23% of participants (>70 years, cognitively intact, well nourished, n = 171) had poor adherence, defined as taking <50% of the prescribed single evening dose of 200 mL of supplement over 60 days in the hospital and community setting. Adherence was evaluated retrospectively at 2 months by caregiver or participant self-report and thus may be an overestimate. Lawson et al11 prescribed short-term (mean 14 days) inpatient supplementation of 200–240 mL (1 carton of juice or milk based supplement, 1260 kJ; 300 kcal) twice daily on the drug cart. Patients (n = 84) were an average 72 (range 40–88) years old, 80% were admitted for elective orthopedic surgery and 60% were overweight. Twenty percent of patients refused to take any supplement, and a further 39% discontinued supplement at some time during their hospital stay. Only 15% of patients took >40% of the total number of supplement doses prescribed twice daily for the duration of their hospital stay. Thus, adherence in this relatively healthy, general orthopedic sample was very poor compared with our frail, elderly, lower limb fracture participants. Bruce et al8 prescribed (on a quasirandomization basis) 1 can of supplement (235 mL, 6.3 kJ; 1.5 kcal/mL) per day to 50 well-nourished (BMI 20–30 kg/m2) women with hip fracture commencing 2–3 days postsurgery and continuing for 28 days in both the inpatient and outpatient setting. The prevalence of cognitive impairment was similar to that in our study. Only 46% of patients consumed 100% and 63% consumed 75% or more of the 28 prescribed cans, even with constant encouragement and variation of flavor and timing of the supplement. The mean number of cans taken was 20.6/28, which represents an overall adherence rate of 73% compared with 67% in our study. It appears that 7/46 (15%) took up to 3 of 28 cans and can be deemed to have refused supplementation, compared with 29% refusal in our study. Counting used or unused cans does not account for partial consumption of cans and may overestimate adherence. Overall, our adherence rates on a refusal or volume basis were similar to or slightly lower than comparable, randomized controlled studies. This is disappointing, given the considerable efforts made to optimize adherence, including individual prescription; administration as a medication; attention to subject preference for flavor, temperature, and scheduling of home intake; extensive encouragement, including triweekly home visits; and detailed monitoring of intake. However, our study prescribed greater supplement volume for a longer period, required administration at home, and participants were selected to be undernourished, with the inclusion of cognitively impaired individuals. Our data suggest that prescription of larger volumes and energy-dense supplements does not compromise adherence and results in net improved energy intake. It is noteworthy that if only those willing to continue to take the supplement are considered, adherence rises to almost 90%. There were no differences in adherence according to gender, admission accommodation, or cognition. This is consistent with the evidence indicating no relationship between medication adherence and gender, education level, socioeconomic status, occupation, marital status, or cognitive function.21,22 However, age appears to influence adherence to nutrition supplements as younger patients consumed a higher proportion of supplements prescribed and the "nonrefusers" were younger. Williams et al10 suggest that the early "refusers" were older, but Bruce et al8 failed to find a difference in age between those who took >75% or <75% of the prescribed number of cans. However, although the latter study included a similar proportion of cognitively impaired patients, and participants had similar baseline BMI to our study, the different volumes and duration of supplementation prescribed and definitions of adherence make comparisons difficult. It is surprising that there was no difference in levels of adherence between those who spent the majority of the intervention period in institutional care (supervised administration via the drug cart) as opposed to the community (self-administration). Furthermore, a higher proportion of those classified as "nonrefusers" were discharged home compared with "refusers." It is possible that administration via the drug cart establishes the supplement as part of medical treatment and encourages intake, an attitude that persists postdischarge. It may also be that once a routine of consumption is established, the level of adherence remains relatively constant. A relatively constant level of supplement intake across the 6 weeks in the "nonrefusers" (Figure 2) supports this. There are no other studies in this patient group to our knowledge that have directly compared adherence according to the method of administration or at the different points of the supplementation period.
The levels of adherence in our study suggest that it is unlikely that
estimated total energy requirements were met. Supplement was prescribed to
provide 45% of BMR plus factors to account for skeletal trauma and activity. A
median adherence level of 67% of the supplement prescription thus
theoretically provides approximately 50% of BMR. If, as consistent with
limited literature estimates of dietary intake in similar patients, dietary
intake is assumed to provide 55% of
BMR,5 then our
results suggest that theoretically overall total energy intake (supplement and
diet) provided approximately 100% of estimated BMR. Thus, estimated total
intake would not cover energy needs for healing, activity, or weight gain,
supported by data from the present study suggesting that only 4 of 41
participants achieved It is acknowledged that in the clinical setting, nutrition support of undernourished patients is likely to include a range of individualized dietetic strategies, including fortification of food, variation of supplement type (energy density and non–milk based), and menu selection, including snacks, to meet patients' preferences. Although such individualized care may improve overall individual adherence and intake, the resulting variation in intervention is difficult to manage within the context of clinical trial design and analysis. Alternatively, adherence may be enhanced within the research context due to increased monitoring and awareness by staff. Continuous evaluation of volitional dietary intake for the duration of supplement prescription was not included in this study due to resource constraints, difficulties with valid assessment of intake in this patient group, particularly at home, and the potential for increased participant and caregiver burden. However, the available (albeit limited) evidence suggests that oral supplements in hip fracture patients do not significantly reduce volitional dietary intake.5–6,23 The finding of weight maintenance in the "nonrefusers" compared with clinically significant weight loss in the "refusers" suggests supplement intake does not displace dietary intake. This study highlights that even with a high level of staff investment and individualized attention, adherence remains a significant problem with potentially important impact on total intake relative to requirements and overall study outcomes. These results support the conclusion by Williams et al10 that staff encouragement was not particularly effective in overcoming poor adherence and that nasogastric feeding should be considered for high-risk patients with poor motivation. Our study suggests older patients (>85 years) are more likely to consume less or refuse supplement and those who refuse supplements early appear unlikely to resume consumption of the supplement. Overall, these data indicate that supervision and coaxing are unlikely to be effective in ensuring adequate intake of oral supplements in at-risk patients such as those >85 years and early "refusers." In order to optimize the capacity for nutrition support to improve clinical outcomes, consideration should be given to providing tube feeding for such patients and others who fail to meet estimated energy needs. Importantly, the use of an energy-dense supplement and prescription of higher volumes has apparently not resulted in substantially lower adherence levels than other studies, although the paucity and limitations of comparable adherence data have already been highlighted. Clearly, similar adherence to a prescription of higher volume of an energy-dense supplement will result in increased net intake, although the effect on volitional intake needs to be verified. Further work is required to understand barriers to adherence, including patient perceptions of need for and benefits of treatment, adverse effects, effect of depression, and the impact on volitional and overall intakes. Achieving estimated energy requirements via oral supplements may be impossible in a substantial proportion of patients, and the need for and outcomes of enteral tube feeding require further investigation. The present study did not collect data on the willingness of "refusers" to accept this more invasive mode of feeding. A pilot study to evaluate the feasibility and acceptability of enteral tube feeding would be an important prerequisite to further studies. Despite the clinical and research importance of adherence, reporting is either absent or incomplete and extremely variable. Poor adherence data, variability in the volume and nutrient composition of supplements, and the absence of rationale for prescription make it difficult to compare studies. Oral supplementation trials should clearly report the underlying rationale for and specific composition and volumes of supplement prescribed, the means of monitoring adherence, the percent of prescription achieved, and actual volume, energy, and nutrient intakes. Data on the extent to which estimated individual energy requirements are met should also be provided. Detailed adherence data and intention-to-treat analysis are essential for valid interpretation and comparison of studies. In the clinical setting, consideration should be given to alternative methods of feeding such as using a feeding tube, particularly in those patients where adherence to oral nutrition supplements is poor and diet alone is insufficient to meet estimated energy requirements. However, expected benefits should be weighed against potential side effects (eg, aspiration, feeding tube obstruction, agitation) and patient acceptability. In addition, ethical and legal issues must be considered, particularly in those with impaired cognition (19 out of 49 in our sample) and decision-making capacity.24,25 There is evidence that tube feeding may be ineffective and counterproductive in patients with significant dementia.26 This study identifies adherence to oral supplement prescription as an important limitation in meeting energy requirements in lower limb fracture patients. Further studies are required on the feasibility and acceptability of alternative approaches to ensuring adequate energy intake and hence optimizing nutrition status and health outcomes. We thank the patients who participated in the trial and members of nursing and residential care staff who administered the intervention. We also acknowledge the contribution of Kylie Lange, Computing Services Flinders University, for statistical support. This research was conducted with support from the following sources: NHMRC Public Health Postgraduate Research Scholarship, Flinders University–Industry Collaborative Research Grant and Nutricia Australia Pty Ltd.
Nutrition in Clinical Practice, Vol. 20, No. 5,
569-578 (2005)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
65 years)
hip-fracture
patients
Top
Materials and Methods
Results
2 test of association as appropriate. Percent of prescribed
volume consumed was compared across admission accommodation, cognition, and
age group (70–74 years; 75+ years) using the Mann-Whitney test. A paired
Wilcoxon test was used to assess the difference in adherence between
supervised and self-administered nutrition supplementation in the 31
participants who received both. The 
=
0.05 (p < .05) was used throughout. 
