This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Improving Food Intake in Nursing Home Residents With Feeding Assistance

A Staffing Analysis

^a Department of Geriatrics, Borun Center for Gerontological Research, University of California School of Medicine, Los Angeles
^b Jewish Home for the Aging, Reseda, California
^c Veterans Administration Greater Los Angeles Healthcare System, Sepulveda Geriatric Research, Education, and Clinical Center, Sepulveda, California

Sandra F. Simmons, Jewish Home for the Aging/UCLA Borun Center for Gerontological Research, 7150 Tampa Avenue, Reseda, CA 91335 E-mail: ssimmons{at}ucla.edu.

Decision Editor: John E. Morley, MB, BCh

	Abstract

Top Abstract Methods Results Discussion References

 
Background. Recommendations have been made to increase the number of nursing home (NH) staff available to provide feeding assistance during mealtime. There are, however, no specific data related to two critical variables necessary to estimate mealtime staffing needs: ⁽¹⁾ How many residents are responsive to feeding assistance? ⁽²⁾ How much staff time is required to provide feeding assistance to these residents? The purpose of this study was to collect preliminary data relevant to these two issues.

Methods. Seventy-four residents in three NHs received a 2-day, or six-meal, trial of one-on-one feeding assistance. Total percentage (0% to 100%) of food and fluid consumed during mealtime was estimated across 3 days during usual NH care and 2 days during the intervention. The amount of time that staff spent providing assistance and type of assistance (i.e., frequency of verbal and physical prompts) was measured under each condition.

Results. One half (50%) of the participants significantly increased their oral food and fluid intake during mealtime. The intervention required significantly more staff time to implement (average of 38 minutes per resident/meal vs 9 minutes rendered by NH staff).

Conclusions. The time required to implement the feeding assistance intervention greatly exceeded the time the nursing staff spent assisting residents in usual mealtime care conditions. These data suggest that it will almost certainly be necessary to both increase staffing levels and to organize staff better to produce higher quality feeding assistance during mealtimes.

UNDERNUTRITION is a common problem among nursing home (NH) residents ^{(1)(2)(3)(4)(5)(6)} that is associated with numerous adverse, costly clinical outcomes ^{(7)(8)(9)(10)(11)}. Physiological (e.g., reduced taste sensations) and medical factors (e.g., dementia, depression) may contribute to undernutrition, but several recent studies suggest that behavioral and environmental factors may be more important determinants of food intake in the NH setting. These factors include: staff failure to identify residents with poor oral intake ^(12)(13)(14); inadequate staff to provide feeding assistance; feeding assistance that fails to promote independence; and social isolation during mealtime (12,^{(15)(16)(17)(18)}). On the basis of these studies, recommendations have been made by multiple groups to increase NH staffing ratios, particularly the number of certified nursing assistants (CNAs), during mealtime ^(16)(19)(20). These staffing recommendations, however, are based mostly on observational data, which indicate that staff do not provide adequate assistance, and on expert consensus regarding optimal feeding assistance care. There are, however, no specific data available related to two critical variables necessary to estimate mealtime staffing needs in NHs. These variables are: (i) how many residents with low intake are responsive to feeding assistance and (ii) how much time is required to provide assistance at mealtimes with those residents who are responsive. We will report preliminary data relevant to these two variables in this article.

	Methods

Top Abstract Methods Results Discussion References

 
Setting and Recruitment
Participants were recruited from three NHs, two of which were proprietary, which housed a total of 482 residents. CNA-level staff-to-resident ratios across the three NHs, as reported by the director of nursing, were typical of industry standards (9 to 10:1 during the 7 AM to 3 PM shift and 12 to 15:1 during the 3 PM to 11 PM shift). Inclusion criteria required residents to be long-stay (non-Medicare), free of a feeding tube, and have problems with low oral intake during mealtimes in the absence of a medical reason that prevented the resident from being able to intake food orally (gastrointestinal or dental disease).

A total of 146 residents in the three homes, who were due a minimum data set (MDS) assessment, were evaluated by research staff over 3 days (nine meals). All residents who ate less than 75% of the food offered during most of these meals were identified as having low intake because the 75% criteria is used on the MDS to identify residents potentially at risk for undernutrition ⁽²¹⁾. The procedure utilized to estimate oral intake involved taking photographs of each participant's meal tray before and after each meal in addition to direct observations during the mealtime period, both of which were conducted by trained research staff. The photography method has been shown to be a reliable and valid method for estimating food and fluid intake during mealtimes for NH residents ⁽¹⁴⁾. Percentage estimates for each food and fluid item and total intake were calculated on the basis of the photographs by different trained research staff who were blind to the purposes of the trial as to whether the participant was in the usual NH care or feeding assistance intervention condition. The agreement between the intake percentage estimates based on the direct observations and the estimates conducted by different blinded research staff members based on the photographs for the participants in this study was high (r = .89, p < .001). Percentage estimates were calculated as opposed to caloric estimates because NH staff, specifically CNAs, are required to conduct daily percentage estimates for each meal, and these daily percentage estimates are used to identify residents who are at risk for undernutrition due to poor oral intake (i.e., consuming less than 75% of most meals).

During each condition, research staff also recorded the frequency and content of verbal (e.g., "how is your lunch today?" or "why don't you try a bite of your soup?") and physical prompts (e.g., physically guiding resident's hand to food or fluid items, helping resident to hold food or fluid items and guide to mouth, feeding resident bites of food) rendered by staff, the total amount of time staff were observed to provide assistance to residents, and the total time that residents had access to their meal trays. In addition, the presence or absence of the following resident behaviors, which might interfere with eating or rendering feeding assistance ^{(22)(23)(24)(25)}, were documented for each meal: refusal of food or fluid items; complaints about taste; health complaints (e.g., pain, nausea); refusal of staff assistance; depression symptoms (e.g., crying); verbal or physical agitation (e.g., repetitive movements or verbalizations) or aggression (i.e., yelling, hitting); swallowing problems (e.g., coughing, drooling); and slow eating pace, defined as requiring greater than 25 minutes to complete a meal ⁽²⁶⁾. Level of agreement between two independent observers in 29 cases for these measures was high (intercorrelations ranged from .88 to .99, p < .001).

A total of 117 (80%) of the 146 residents were identified as having low intake, which is similar to the proportions of residents reported in other studies to have low intake on the basis of the 75% MDS criteria ^(12)(13)(14). Of these 117 residents, a total of 20 (17%) were ineligible for this study due to transfer ⁽⁶⁾, feeding tube insertion ⁽²⁾, acute medical problems ⁽⁵⁾, or death ⁽⁷⁾. Research staff attempted to recruit the remaining 97 eligible residents. Consents were obtained from 74 participants (76%) of this eligible group.

Measures
Demographic, medical, and functional information was retrieved from each participant's medical record. Each participant's height was obtained from the chart, and independent assessments of body weight were conducted by research staff to calculate body mass index (BMI) values. The most recent MDS assessments (i.e., annual and quarterly) were retrieved from each participant's medical chart. The MDS-derived Cognitive Performance Scale (CPS) score was calculated for each participant ⁽²⁷⁾. The CPS total score ranges from 0 (cognitively intact) to 6 (severely impaired).

Feeding Assistance Intervention Protocol
Participants received a 2-day or six-meal (i.e., breakfast, lunch, and dinner on two consecutive days) trial of feeding assistance implemented by trained, CNA-level, research staff. The intervention consisted of the following components: one-to-one, continuous assistance; a standardized graduated prompting protocol that enhanced self-feeding capabilities; social interaction throughout the mealtime period; compliance with resident preference for dining location and type of assistance; proper positioning for eating; meal tray substitutions if preferred by a resident and extended access to tray up to 1.5 hours per meal. These intervention components have been identified as defining optimal feeding assistance quality ^{(28)(29)(30)(31)(32)}. All observation and photographic food intake measures that were completed during the usual care condition were repeated by research staff during the intervention condition.

A subsample of nine participants, who did not increase their intake in response to the initial 2-day intervention, received two to four additional intervention days to evaluate the accuracy of a 2-day intervention trial for determining a resident's responsiveness to feeding assistance. A different subsample of 18 participants, who showed significant increases in intake in response to the initial 2-day intervention, received a second 2-day trial during which the intervention was delivered to small groups of three residents to evaluate the feasibility of a group intervention.

	Results

Top Abstract Methods Results Discussion References

 
Participants and Setting
Table 1 shows the demographic, medical, and nutritional characteristics of the 74 intervention participants. The participants were typical of a community-dwelling, long-stay NH population in that they were predominately female (89%) and white (95%). They were moderately cognitively impaired as indicated by the prevalence of physician-recorded chart diagnoses of dementia (46%) and the MDS-derived CPS score (3.1 ± 1.6). The prevalence of physician-recorded chart diagnoses of depression was 31%. The BMI of participants showed that 43% ⁽³²⁾ had values indicative of undernutrition (BMI < 20). The majority (72%) of the participants were on a therapeutic or mechanically altered diet and/or had chart orders to receive a daily nutritional supplement (81%). There were no significant differences on any of these demographic, medical, or nutritional characteristics between the 74 intervention participants and those who did not participate in this study (n = 72) for whatever reason (i.e., adequate intake, no consent, transfer, medical reason, death).

Table 2 displays the results for those 37 participants who showed an average gain in intake of >10% (i.e., 8 "marginally responsive" plus 29 "responsive"). These participants had a significant increase in their intake (47% to 70%; t = 12.86, p < .001); however, the total amount of assistance time was also significantly greater (average of 9 to 38 minutes/resident/meal; t = -12.68, p < .001) as was tray access time (32 to 39 minutes/resident/meal; t = 3.17, p < .01). The quality of assistance also improved as documented by a significant increase in the frequency of verbal prompts to promote independence (2 to 15/resident/meal; t = 7.6, p < .001).

Table 3 displays the same results for the 37 participants who showed little to no change in their intake (46% to 47%), despite the significant increases in total assistance time (3 to 39 minutes/resident/meal; t = -20.12, p < .001), frequency of verbal prompts (1 to 12/resident/meal, t = 7.39, p < .001), and extended tray access time (35 to 39 minutes/resident/meal; t = 2.12, p < .05).

There was little variability in participants' responses (i.e., average total intake) to the intervention between the six meals or the 2 days (intercorrelations ranged from .83 to .94, p < .001). Thus, participants' classification as "responsive" (i.e., intake gains 10%) or "unresponsive" (i.e., intake gains <10%) based on one (i.e., three meals) versus 2 days (i.e., six meals) of the intervention remained the same for the majority (87%, = .73, p < .001). Similarly, the classification of all nine participants, who were classified as unresponsive based on the 2-day trial, remained unresponsive based on 4 to 6 intervention days (average intake in usual care, 2-day trial, vs additional days, 49%, 57%, and 55%, respectively).

An exploratory correlational analysis in which all variables listed in Table 1 , in addition to resident behaviors exhibited in the context of the intervention, showed the following seven variables to be significantly intercorrelated with change in the average total percentage intake: length of stay; number of diagnoses; CPS total score; total number of intervention meals (i.e., out of a possible total of six) during which the participant expressed food complaints; food refusal; symptoms of depression; or a slow eating pace. The intercorrelations ranged from -.24 (p < .05) to .50 (p < .001). These seven variables were entered into a regression analysis to predict change in average total percentage intake between the two conditions: usual care and the intervention. The model was significant (R = .66, F = 7.34, p <.001), with CPS total score (t = 4.08, p < .001) and slow eating pace (t = -2.40, p < .05) emerging as the two significant variables in the equation. A higher level of cognitive impairment was associated with a greater increase in intake, while a slow eating pace was associated with a smaller increase in intake.

	Discussion

Top Abstract Methods Results Discussion References

 
The time requirements to implement the feeding assistance intervention greatly exceeded the time NH staff were directly observed to spend assisting residents in usual mealtime care. These data suggest that it will almost certainly be necessary to both increase staffing levels and to organize staff better to improve the quality of feeding assistance during mealtimes, which has been suggested by multiple groups ^(16)(19)(20). These staffing investments, however, could be most effectively utilized if several additional factors are considered.

First, and most important, residents who are not responsive to feeding assistance during mealtimes must be correctly identified and alternative interventions developed for this group. The failure to identify these unresponsive residents would roughly double the number of staff needed during mealtime because, on the basis of this study, approximately 50% are estimated to be unresponsive. In addition, it would be difficult to manage staff since they would be expected to provide feeding assistance to a frustrating mixture of residents, some of whom respond well to assistance and some of whom do not.

Residents who were more severely cognitively impaired and, thus, required a greater amount of physical assistance to eat were more likely to be responsive to the intervention. In contrast, unresponsive residents were likely to be less cognitively impaired and able to eat independently and preferred to do so (i.e., accepting mostly verbal cueing from staff but refusing physical assistance) but ate slowly. The extended tray access time did not increase intake in this group nor did the social interaction and verbal cueing provided by research staff. Our preliminary data, however, strongly suggest that a 1- to 2-day trial of feeding assistance may be the most efficient and valid method of identifying residents who are unresponsive to assistance during mealtimes as opposed to relying on any resident characteristics. The strong associations between performance on day 1 versus day 2 of the feeding assistance trial and performance over longer time periods (i.e., days 3 through 6) that we report is encouraging; but, given the importance of correctly identifying unresponsive residents, it is clearly necessary to replicate these findings.

Second, the staff requirements to provide feeding assistance could be reduced further if staff were able to effectively provide feeding assistance to multiple residents simultaneously as opposed to individually. Again, our preliminary data suggest that feeding assistance can be effectively provided in small groups for most residents who are responsive to individual assistance, but additional time is required to get residents to and from the dining room for each meal. On the basis of our direct observations of usual NH care in this study, many residents were fed in their rooms during breakfast and dinner because of the time requirements to get these residents out of bed, dressed, and into the dining room. It is important to note that our travel time estimates, which ranged from an additional 1 to 10 minutes per resident per meal, did not include the time required to get residents out of bed, dressed, and groomed in preparation for transport to the dining room, all of which would substantially increase the minimal time required for the breakfast mealtime period.

Finally, food complaints and refusal of food were common among both responsive and unresponsive participants, although none of the participants were identified by NH staff as having food complaints according to MDS documentation. Food complaints may stem from a combination of poor food quality, diet orders, or lack of appetite associated with depression, medical conditions, and/or medications that have anorexigenic side effects. Alternative and supplemental interventions for all NH residents thus should address issues such as food quality, the necessity of diet and anorexigenic medication orders, and symptoms of depression that may not be treated adequately within nursing homes ^(33)(34)(35).

In conclusion, the failure to accurately identify residents who are unresponsive to mealtime feeding assistance will reduce the effectiveness of any intervention that simply increases mealtime staffing under the false assumption that all residents will improve intake with enough staff attention. The failure of comprehensive, individualized, one-on-one feeding assistance to improve intake among a significant portion of residents in this study suggests that this is not the case and, more importantly, highlights the fact that the relationship between staffing and residents' food intake is complex.

	Acknowledgments

 
This research was supported by the University of California, Los Angeles, Older Adults Independence Center through a Career Development Award, and the Beverly Manor Corporation. We thank David Reuben for his thoughtful review of this manuscript prior to submission. We also thank all of the research assistants at the Borun Center for their assistance with data collection; a special thanks goes to Karen Glienke for her instrumental role in the success of this project.

Received February 6, 2001

Accepted February 14, 2001

	References

Top Abstract Methods Results Discussion References

 

1. Rudman D, Feller AG, 1989. Protein-calorie undernutrition in the nursing home. J Am Geriatr Soc. 37:173-183. [Medline]
2. Keller HH, 1993. Malnutrition in institutionalized elderly: How and why?. J Am Geriatr Soc. 41:1212-1218. [Medline]
3. Abbasi AA, Rudman D, 1993. Observations on the prevalence of protein-calories undernutrition in VA nursing homes. J Am Geriatr Soc. 41:117-121. [Medline]
4. Morley JE, Kruenzle D, 1994. Causes of weight loss in a community nursing home. J Am Geriatr Soc. 42:583-585. [Medline]
5. Blaum CS, Fries BE, Fiatarone MA, 1995. Factors associated with low body mass index and weight loss in nursing home residents. J Gerontol Med Sci. 50A:M162-M168. [Abstract]
6. Report to the Special Committee on Aging, U.S. Senate. California Nursing Homes Care Problems Persist Despite Federal and State Oversight. GAO/HEHS-98–202. Washington, DC: United States General Accounting Office; 1998.
7. Bergstrom N, Braden B, 1992. A prospective study of pressure sore risk among institutionalized elderly. J Am Geriatr Soc. 40:747-758. [Medline]
8. Dwyer JT, Coleman KA, Krall E, et al. 1987. Changes in relative weight among institutionalized elderly adults. J Gerontol. 42:246-251. [Abstract/Free Full Text]
9. Ferguson RP, O'Conner P, Crabtree B, Batchelor A, Mitchell J, Coppola D, 1993. Serum albumin and realbumin as predictors of clinical outcomes of hospitalized elderly nursing home residents. J Am Geriatr Soc. 41:545-549. [Medline]
10. Rudman D, Mattson DE, Feller AG, Nagraj HS, 1989. A mortality risk index for men in a Veterans Administration extended care facility. J Parenter and Enteral Nutr. 13:189-195.
11. Woo J, Chan SM, Mak YT, Swaminathan R, 1989. Biochemical predictors of short term mortality in elderly residents of chronic care institutions. J Clin Pathol. 42:1241-1245. [Abstract/Free Full Text]
12. Kayser-Jones J, Schell E, Porter C, Paul S, 1997. Reliability of percentage figures used to record the dietary intake of nursing home residents. Nurs Home Med. 5:69-76.
13. Pokrywka HS, Koffler KH, Remsburg R, et al. 1997. Accuracy of patient care staff in estimating and documenting meal intake of nursing home residents. J Am Geriatr Soc. 45:1223-1227. [Medline]
14. Simmons SF, Reuben D, 2000. Nutritional intake monitoring for nursing home residents: a comparison of staff documentation, direct observation, and photography methods. J Am Geriatr Soc. 48:209-213. [Medline]
15. Kayser-Jones J, 1996. Mealtime in nursing homes: the importance of individualized care. J Gerontol Nurs. 3:26-31.
16. Kayser-Jones J, Schell E, 1997. The effect of staffing on the quality of care at mealtime. Nurs Outlook. 45:64-72. [Medline]
17. Amella EJ. Factors influencing the proportion of food consumed by nursing home residents with dementia. Advancing Geriatrics Nursing Practice. Mezey M, Fulmer T, series eds. J Am Geriatr Soc. 1999;47:879–885.
18. Schell ES, Kayser-Jones J, 1999. The effect of role-taking ability on caregiver-resident mealtime interaction. Appl Nurs Res. 12:38-44. [Medline]
19. Mondoux L. Testimony of the American Nurses Association before the National Academy of Sciences Institute of Medicine Committee on Improving Quality in Long Term Care. Washington, DC: National Academy Press; 1998.
20. Schnelle JF, Cretin S, Saliba D, Simmons SF. Minimum Nurse Aide Staffing Required to Implement Best Practice Care in Nursing Homes. Chapter in report to Congress: Appropriateness of minimum nurse staffing ratios in nursing homes. Washington, DC: Health Care Financing Administration. Summer 2000: Vol. 2, chapter 14.
21. Long Term Care Facility Resident Assessment Instrument (RAI) User's Manual. Natick, MA: Eliot Press; April 1999.
22. Durnbaugh T, Haley B, Roberts S, 1996. Assessing problem feeding behaviors in mid-stage Alzheimer's disease. Geriatr Nurs. 17:63-67. [Medline]
23. Kline N, Sexton DL. Eating behaviors of nursing home residents who display agitation. Nurs Manage. 1996;27(9):32JJ-LL–32PP.
24. Steele CM, Greenwood C, Ens I, Robertson C, Seidman-Carlson R, 1997. Mealtime difficulties in a home for the aged: not just dysphagia. Dysphagia. 12:43-50. [Medline]
25. Tully MW, Matrakas KL, Muir J, Musallam K, 1997. The Eating Behavior Scale: a simple method of assessing functional ability in patients with Alzheimer's disease. J Gerontol Nurs. 7:9-15.
26. Keller HH, 1993. Malnutrition in institutionalized elderly: How and why?. J Am Geriatr Soc. 41:1212-1218.
27. Hartmaier SL, Sloane PD, Guess HA, Koch GG, Mitchell CM, Phillips CD, 1995. Validation of the Minimum Data Set cognitive performance scale: agreement with the Mini-Mental State Examination. J Gerontol Med Sci. 50A:M128-M133. [Abstract]
28. Lange-Alberts ME, Shott S, 1994. Nutritional intake: use of touch and verbal cuing. J Gerontol Nurs. 2:36-40.
29. Manning AM, Means JG, 1975. A self-feeding program for geriatric patients in a skilled nursing facility. J Am Diet Assoc. 66: (3) 275-276. [Medline]
30. Musson ND, Kincaid J, Ryan P, et al. 1990. Nature, nurture, nutrition: interdisciplinary programs to address the prevention of malnutrition and dehydration. Dysphagia 5:96-101. [Medline]
31. Osborn CL, Marshall M, 1992. Promoting mealtime independence. Geriatr Nurs. 13:254-256. [Medline]
32. Van Ort S, Phillips LR, 1995. Nursing interventions to promote functional feeding. J Gerontol Nurs. 10:6-14.
33. Buckler DA, Kelber ST, Goodwin JS, 1994. The use of dietary restrictions in malnourished nursing home patients. J Am Gerontol Soc. 42:1100-1102. [Medline]
34. Morley JE, Kraenzle D, 1994. Causes of weight loss in a community nursing home. J Am Gerontol Soc. 42:583-585.
35. Morley JE, Silver AJ, 1995. Nutritional issues in nursing home care. Ann Intern Med. 123:850-859. [Abstract/Free Full Text]

This article has been cited by other articles:

				I. Gasquet, S. Dehe, P. Gaudebout, and B. Falissard Regular Visitors Are Not Good Substitutes for Assessment of Elderly Patient Satisfaction With Nursing Home Care and Services J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2003; 58(11): M1036 - 1041. [Abstract] [Full Text] [PDF]

				M.-M. G. Wilson and J. E. Morley Invited Review: Aging and energy balance J Appl Physiol, October 1, 2003; 95(4): 1728 - 1736. [Abstract] [Full Text] [PDF]

				J. E. Morley, J. H. Flaherty, and D. R. Thomas Editorial: Geriatricians, Continuous Quality Improvement, and Improved Care for Older Persons J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2003; 58(9): M809 - 812. [Full Text] [PDF]

				D. Tsilimingras, A. K. Rosen, and D. R. Berlowitz Review Article: Patient Safety in Geriatrics: A Call for Action J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2003; 58(9): M813 - 819. [Abstract] [Full Text] [PDF]

				L. Volicer and A. C. Hurley Review Article: Management of Behavioral Symptoms in Progressive Degenerative Dementias J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2003; 58(9): M837 - 845. [Abstract] [Full Text] [PDF]

				W. A. Banks and J. E. Morley Memories Are Made of This: Recent Advances in Understanding Cognitive Impairments and Dementia J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2003; 58(4): M314 - 321. [Full Text] [PDF]

				J. E. Morley Anorexia and Weight Loss in Older Persons J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2003; 58(2): M131 - 137. [Full Text] [PDF]

				C. Paquet, D. St-Arnaud-McKenzie, M.-J. Kergoat, G. Ferland, and L. Dube Direct and Indirect Effects of Everyday Emotions on Food Intake of Elderly Patients in Institutions J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2003; 58(2): M153 - 158. [Abstract] [Full Text] [PDF]

				N. L. Crogan and A. Pasvogel The Influence of Protein-Calorie Malnutrition on Quality of Life in Nursing Homes J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2003; 58(2): M159 - 164. [Abstract] [Full Text] [PDF]

				J. E. Morley Editorial: Citations, Impact Factor, and the Journal J. Gerontol. A Biol. Sci. Med. Sci., December 1, 2002; 57(12): M765 - 769. [Full Text] [PDF]

				S. F. Simmons, S. Babineau, E. Garcia, and J. F. Schnelle Quality Assessment in Nursing Homes by Systematic Direct Observation: Feeding Assistance J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2002; 57(10): M665 - 671. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation