|Home | About | Journals | Submit | Contact Us | Français|
There is little evidence regarding the type(s) of information clinicians use to make the recommendation for oral or nonoral feeding in patients with oropharyngeal dysphagia. This study represents a first step toward identifying data used by clinicians to make this recommendation and how clinical experience may affect the recommendation. Thirteen variables were considered critical in making the oral vs. nonoral decision by the 23 clinicians working in dysphagia. These variables were then used by the clinicians to independently recommend oral vs. nonoral feeding or partial oral with nonoral feeding for the 20 anonymous patients whose modified barium swallows were sent on a videotape to each clinician. Clinicians also received data on the 13 variables for each patient. Results of clinician agreement on the recommendation of full oral and nonoral only were quite high, as measured by Kappa statistics. In an analysis of which of the 13 criteria clinicians used in making their recommendations, amount of aspiration was the criterion with the highest frequency. Recommendations for use of postures and maneuvers and the effect of clinician experience on these choices were also analyzed.
One of the most important recommendations made by clinicians caring for oropharyngeal dysphagic patients is whether the patient should receive oral nutrition. The recommendation is usually made by a speech-language pathologist based upon the results of a patient evaluation which may include a radiographic study, the modified barium swallow , an endoscopic study, and/or a bedside evaluation.
Despite the importance of this recommendation to most dysphagic patients and their quality of life, there is little evidence as to the type(s) of information clinicians use to make this or other recommendations . Usually, information about frequency and amount of aspiration is used in this decision-making about introduction of oral intake [3–8]. Recently, the presence of penetration (entry of food or liquid into the airway and out again during a swallow viewed radiographically) has been used as a rationale for the nonoral feeding decision by some clinicians . Other data are likely also used to make the nonoral feeding recommendation but have never been defined. No written guidelines exist for making this decision. This study represents a first step toward identifying the data used by clinicians to make the nonoral/oral recommendation for dysphagic patients. In this study, 23 clinicians were involved in identifying the information they typically used to make this recommendation. There were two parts to the study. Part I identified the important variables and Part II assessed these variables in the clinical setting. This project was approved by the Northwestern University Institutional Review Board.
In Part I several small groups of clinicians participated in one of four think-tank sessions to discuss the information they used in making the decision to recommend oral or nonoral nutrition. These discussions were held in 1-h telephone calls where clinicians were encouraged to talk freely about the types of information they utilized to make this recommendation. At the end of four 1-h telephone calls, there were a total of 46 variables identified by clinicians as variables they might use in making the recommendation. Table 1 presents these 46 variables. The list of 46 variables then was sent to the clinicians to rank the top ten variables that were most important in their decision-making with 1 being most important and 10 being least important. Based upon these judgments, the 13 variables that were considered most critical by the clinicians were identified (there were four variables ranked as 10 bringing the total to 13.) These were (1) amount of aspiration, (2) acuity of medical condition, (3) history of pneumonia, (4) alertness, (5) cough ability, (6) frequency of aspiration, (7) respiratory status, (8) patient wishes, (9) secretion management, (10) silent aspiration, (11) recovery prognosis, (12) ability to complete postures/maneuvers, and (13) diagnosis.
In Part II a videotape containing 20 videofluorographic studies of dysphagic patients, representing a range of severities, ages, and diagnoses, and several normal subjects were copied onto a master file and sent to each clinician along with a short description of each patient's medical history and etiology for aspiration or dysphagia. On average, these gave only the most basic information. Between the information shown on the videotape and the short discussion of the patient, the clinician received data on the 13 variables for each patient. Each clinician was then asked to integrate all of this information to judge each of the 20 subjects as to whether they would recommend each individual become oral, partially oral with nonoral feeding, or nonoral. In this way, we could examine the level of agreement of the 23 clinicians in making the judgment of oral intake status and define the information they used to make their recommendation as done in normal clinical care. Part II data consisted of clinician characteristics and clinician assessments of the 20 patients. Clinicians were also asked what diet consistency(ies) applied if they recommended limited oral, whether dental status affected their decision, and whether they would recommend postures or maneuvers.
Descriptive statistics (mean, median, standard deviation, minimum, maximum) are provided for the clinician characteristics.
For the clinician assessments, frequencies and percents are given for all items. Multirater Kappa statistics  are given to determine the amount of agreement among the 23 clinicians. Kappa values near 0 indicate no agreement. Kappa values near 1 indicate agreement. Agreement is calculated by looking at the 23 clinician ratings for one patient and getting a measure of how different these ratings are. These difference measures are then totaled over the 20 patients. If there is a lot of difference in clinician ratings stratified by patient, then Kappa will be low. Conversely, if there is no difference in clinician ratings, Kappa will be high. Landis and Koch  relate strength of agreement to the Kappa statistic as follows:
In the tables giving the statistics for the clinician assessments, a Kappa statistic together with its qualitative strength rating is given for each category of response. Kappa statistics are corrected for chance agreement. In relating clinician characteristics to clinician assessments, the Mantel-Haenszel test  is used by calculating a statistic that is stratified by patient.
Table 2 presents the characteristics of the 23 participating clinicians, including their years of experience, numbers of videofluorographic studies completed per month in the last year, and the number of years they have spent in treating dysphagic patients. The typical clinician in this study had a mean of 14.6 years of experience in swallowing and conducts an average of 20 radiographic studies per month.
When patient recommendations for oral or nonoral feeding as made by clinicians were examined, clinicians could answer with full oral, limited oral, nonoral only, could not decide, or did not answer. Clinicians were also asked to circle as many of the 13 criteria that they used in making each of their recommendations regarding oral vs. nonoral feeding for each of the 20 patients. We examined how frequently each criterion was used and whether there was agreement among clinicians in the use of the criteria (Table 3). The amount of aspiration was the criterion with the highest frequency used among clinicians. The next two most frequently used criteria were frequency of aspiration and medical diagnosis. The history of pneumonia was used in 40% of the assessments with a rank of 4 among the 13 criteria and had the best agreement.
Table 4 presents the clinician agreement on recommendations. There was substantial agreement on the full oral or nonoral recommendation but not as much for limited oral. Dental status was not a factor in the recommendations. There was fair agreement on whether a posture or maneuver was recommended (Table 4).
Table 5 presents the frequency with which the various recommendations for use of specific postures and maneuvers were made and the order of their recommendation. A wide range of postures and maneuvers was suggested, although effortful swallow, head turn, chin tuck, super supraglottic swallow, supraglottic swallow, and repeat swallow were most often recommended. Recommendations are classified according to the order in which clinicians reported them (first response, second response, third response, fourth response).
Experience made a difference in clinician classification of oral status (Table 6). Full oral feeding was recommended at the same rate regardless of experience. However, more experienced clinicians recommended limited oral intake more often than nonoral and less experienced clinicians did the opposite. Recommendations regarding postures or maneuvers also varied with years of experience. More experienced clinicians tended to recommend a posture or maneuver less often than less experienced clinicians.
This study provided no training for making the decision regarding oral, limited oral, or nonoral feeding. No profession involved in swallowing provides such regular training. Considering that there is little formal clinician training available regarding the criteria used to make the overall recommendation for oral, partial oral with nonoral, or nonoral feeding of a patient, the level of agreement between clinicians seen in this study is surprisingly good. However, when including the variable “limited oral,” the experience of the clinicians became important in making the recommendation. Data indicate that less experienced clinicians (<15 years) are less likely to recommend limited oral compared with more experienced clinicians (>15 years or more). It appears that more experienced clinicians will tolerate oral intake in a patient who may aspirate a small amount consistently or exhibit a small amount of residue in the pharynx. This may result from the experienced clinician's observations that many such patients do not develop medical problems as a result. Interestingly, less experienced clinicians recommend more swallow maneuvers and postures than more experienced clinicians, perhaps because the less experienced clinicians have more difficulty selecting the most appropriate posture or maneuver for each patient or perhaps more experienced clinicians have found variable compliance by patients in using these techniques. More experienced clinicians may be more familiar with the risk of the possibility that patients sometimes become complacent about their swallowing and stop using a posture or maneuver at meal time because they think their swallowing has recovered. Less experienced clinicians may be more familiar with swallow maneuvers and postures from their more recent educational experiences than more experienced clinicians who may not have learned as much about using those treatment procedures.
This pilot study indicates the need for systematic training of clinicians in making oral, partial oral, or nonoral recommendation(s). Additional data are needed regarding the information that clinicians use in making this recommendation. These data may vary by medical diagnosis, age, gender, and other characteristics of the patient. The present study did not include enough patient examples to permit such an item-by-item analysis. Further research is also needed regarding data most important in predicting patient success with oral or partial oral feeding.
Jeri A. Logemann, Department of Communication Sciences and Disorders, Northwestern University, Frances Searle Building, 2240 Campus Drive, Evanston, Illinois 60208, USA, Northwestern University, 2240 North Campus Drive, Room 3-358, Evanston, IL 60208, USA.
Alfred Rademaker, Department of Preventive Medicine, Robert H. Lurie Comprehensive Cancer Center, Northwestern University, Chicago, Illinois 60611, USA.
Barbara Roa Pauloski, Department of Communication Sciences and Disorders, Northwestern University, Frances Searle Building, 2240 Campus Drive, Evanston, Illinois 60208, USA.
Jodi Antinoja, Froedtert Memorial Hospital, 9200 W. Wisconsin Avenue, Milwaukee, Wisconsin 53226, USA.
Mary Bacon, Audiology and Speech Pathology (126), Hines VA Medical Center, Building 1, Room E-260, Hines, Illinois 60141, USA.
Michelle Bernstein, Department of Otolaryngology, University of Miami Hospital Medical Center, UHMC Room 4025, 1475 12th Avenue NW, Miami, Florida 33136, USA.
Joy Gaziano, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA.
Barbara Grande, St. Joseph Regional Medical Center, 5000 W. Chambers Street, Milwaukee, Wisconsin 53210, USA.
Lisa Kelchner, CAHS-Communication Sciences and Disorders, University of Cincinnati, French – East, P.O. Box 670379, Cincinnati, Ohio 45267-0379, USA.
Amy Kelly, Speech Pathology Department, Evanston Northwestern Healthcare, 2650 Ridge Avenue, Room 2206, Evanston, Illinois 60201, USA.
Bernice Klaben, Blaine Block Institute for Voice Analysis and Rehabilitation, 369 W. First Street, Suite 408, Dayton, Ohio 45402-3065, USA.
Donna Lundy, Department of Otolaryngology, University of Miami Hospital Medical Center, UHMC Room 4025, 1475 12th Avenue NW, Miami, Florida 33136, USA.
Lisa Newman, Army Audiology & Speech Center, Walter Reed Army Medical Center, 6900 Georgia Avenue, Washington, DC 20307-5001, USA.
Daphne Santa, Audiology and Speech (126), Miami VA Medical Center, 1201 NW 16th Street, Miami, Florida 33125, USA.
Linda Stachowiak, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA.
Carrie Stangl-McBreen, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.
Cory Atkinson, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.
Heidi Bassani, Army Audiology & Speech Center, Walter Reed Army Medical Center, 6900 Georgia Avenue, Washington, DC 20307-5001, USA.
Melissa Czapla, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA.
Julie Farquharson, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.
Kristin Larsen, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.
Vicki Lewis, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA.
Heather Logan, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.
Teri Nitschke, Zablocki VA Medical Center, 5000 National Avenue, Milwaukee, Wisconsin 53295, USA.
Sharon Veis, Northwestern Memorial Hospital, 201 E. Huron, Suite 10-205, Chicago, Illinois 60611, USA.