Potvin 2007 Modified from Law, Stewart, Pollock, Letts, Bosch, & Westmorland, 1998

Review 1

Critical Review Form

Quantitative Studies

Reference:

Clawson, E., Kuchinski, K.S., & Bach, R. (2007). Use of behavioral interventions and parent education to address feeding difficulties in young children with spastic diplegic cerebral palsy. NeuroRehabilitation, 22, pp. 397-406.
STUDY PURPOSE:
Was the purpose stated clearly?
_x__ Yes
___ No / Outline the purpose of the study (i.e., study objective or aim):
Assess the effectiveness of an intensive day patient pediatric feeding program using oral motor exercise, behavioral interventions, parental education to increase the oral feeding and nutrition of children with spastic diplegic CP.
LITERATURE:
Was relevant background
literature reviewed?
__x Yes
___ No / Describe the justification of the need for this study (3-4 key points)
Þ Dysphagia is common in children who have mod-severe neurological impairment
1. Feeding is a complicated task, requiring coordination and integration of sensorimotor pathways.
2. Oral dysphagia, pharyngeal dysphagia and/ or motor dysfunction contribute to poor nutritional status of children with CP.
3. Under nutrition causes high surgical morbidity, decreased linear growth, and delayed healing. Poor nutrition effects motor, cognitive and dev. delays.
4. Following nutritional rehab, many children are less irritable, more alert and make progress.
DESIGN:
___ randomized
___ cohort (population -based)
__x_ before and after
___ case-control
___ cross-sectional
(1+ group at 1 point in time)
___ single case design
___ case study / Describe the study design:
Before and after design: Used to eval. a group of clients involved in a treatment. Information is collected about the initial status in terms of the outcomes of interest and then collects info again about the outcomes after treatment is received. This design is used when the evaluator does not wish to withhold treatment from any client. Given that there is no control group, cannot judge if the treatment alone was responsible for any changes. Changes could be due to other factors, incl. med. use, lifestyle or environmental changes
Can the author answer the study question with the study design?
Not completely- it will be more difficult to determine whether the change is due to tx. alone since there is no control group or randomization.
Were the design and/or method used introducing biases. If so describe:
POSSIBLE intervention biases:
It is unclear whether the subjects received any other tx. at the same time (nothing to this effect was stated…given that it was a day tx. program, it is quite possible that they weren’t receiving other feeding-related tx, but it is unclear). It is also unclear whether the results could be related to maturation alone since results were not reported using statistical significance. Therapist involvement was also unclear. It appears that there were teams of therapist-parent-child. It is unclear how many therapists were involved- perhaps 1 OTR, 1 SLP. Who did the exercises and who fed the children is unclear – one therapist could have done it differently than another and it doesn’t seem that this was measured/evaluated.
SAMPLE SIZE:
N = 8
Was sample size justified?
___ Yes
__x_ No
___ N/A
Was Power Discussed?
___ Yes
__x_ No
___ N/A / Sample Description (e.g., age, gender, diagnosis, other characteristics)
4B,4G. mean age 2.8 yrs (1.5 to 4.7yrs), dx: all with Spastic diplegic cerebral palsy (SDCP), 88% oral dysphagia. 5-unable to feed themselves, 3 transitioning to self feeding, 3 g-tube, 1-ng tube, 4-required no supplemental feeding. Criteria for admission inc. children with mod-severe feeding difficulties who were med. stable.
How was sample identified? Was it a representative sample?
Referred by their pediatrician, due to feeding difficulties. Recommendations were based on screening of the ID Team, feeding skills, nutritional status, med. stability. The subjects were children with SDCP, who were admitted to the day feeding program. Average LOS was 29 days. This sample is not representative of the general sample of children with CP since they have chosen a small subgroup.
If there were more than one group, was there similarity and differences between the groups? Describe:
There was only one group
Was informed consent and assent obtained?
Not stated, since parents were an integral part of the study, it is assumed that they approved.
OUTCOMES:
Specify the frequency of outcome measurement (i.e., pre, post, follow-up): Baseline sessions-anthropometric data, caregiver instruction; parent and therapist sessions and the Beckman Oral Motor Assessment was administered, reinforcer eval to id motivational toys/videos, therapist did seating assessment. Measures then taken at every feeding session. Follow-up appts at 1, 4, 7, 12 months following discharge from the day feeding program.
Outcome areas
(e.g., self care, productivity)
Þ Child feeding behaviors- positive and negative (which were defined)
Þ Caregiver feeding behaviors
Anthropometric Data:
Þ Weight
Þ Height
Þ Percent ideal body weight
Þ Expected height and weight
Þ Average calories and volume consumed by mouth and via tube feeds.
Þ Feeding skills
Þ Motivators
Þ Postural needs/activity level / List measures used
(e.g., Sensory Profile, VMI)
Þ Behaviors recorded
Þ Scoring of ability to correctly provide instructions, prompts, and consequences (IPC).
Þ Entered into Center for Disease Control and Prevention anthropometric software program to obtain exact percentiles.
Þ Measured in kilograms using an infant Health-O-Meter bucket scale.
Þ Measured in centimeters using an Infantometers height board.
Þ NCHS growth charts (at 50th percentile weight for height)
Þ Calculated based on child’s age and sex based on research by Foman (who used one reference boy and girl who were likely typically developing).
Weight & height gain/day calculated by: subtract weight /height at previous visit from weight/height at later visit and divide by # of elapsed days between visits. This was then compared to expected data. Actual weight/height gained/day divided by expected value to obtain percentage.
Þ Food weighed at start & end of each meal. Total grams consumed= (pre-weight) – (post-weight - spills or emesis). Total calories consumed calculated using FoodPro computer program.
Þ Clinical observation: Caregiver/therapist fed child without using any therapeutic strategies for 2 days.
Þ Beckman Oral Motor Assessment
Þ Reinforcer evaluation (of toys and videos) in order to reward them during meals.
Seating assessment- to determine best support for trunk and head, positions to decrease spasticity and keep head at midline. / Reliable and Valid?
Þ Not reliable- although behaviors were defined, I wouldn’t say they were fully and accurately measurable. It does not say they measured agreement between recorders and from one session to another with the same child/parent. It says in one place that the “same staff members” conducted the meals throughout admission and in another place says “the therapists” remained the primary feeders until ~2 weeks prior to d/c when the caregivers transitioned into meals. It is unclear how many “feeders” or staff members there were and who these people were.
Þ I assume that this program is for typically developing children, not children with special health needs – However perhaps since all children were measured on the same “inappropriate scale”, it would be valid.
Þ Yes, if the same scale was used, however this may not have been the most accurate/sophisticated tool to use, per the information on their web site.
Þ This was fine assuming all children were under the max. measurement taken by this board, however, spasticity would certainly be one major factor affecting accurate height measurement from measure-to-measure.
Þ This chart is for typically developing children, not children with special health needs. However, if they are only looking at the % change toward more “ideal” body weight, then perhaps this “inappropriate” measure would be fine. Not clear exactly what valuable info. it would share, though.
Þ Unsure how reliable/valid Foman’s research is. It seems that it contained a great amt. of inference/assumptions.
Þ Measuring food intake is not reliable – it is too difficult to get all food that comes back out of mouth and when it comes back out, it can have saliva added, etc. Since calories were calculated based on volume, which would not be truly accurate either.
Þ Could not find reliability/validity data on this assessment. It was not used, however, to measure changes, but rather just to gather information.
Þ Unclear what kind of evaluation they used to determine this.
Available seating- high chairs, toddler chairs, giraffe chairs, and the Tripp Trapp chair. Perhaps these weren’t the most supportive chairs?
Þ
INTERVENTION:
Intervention was described in detail?
___ Yes
__x_ No-unable to replicate
___ Not addressed
Contamination was avoided?
___ Yes
___ No
__x_ Not addressed / Provide a short description of the intervention including type of intervention, who delivered it, how often and in what setting.
Day program staff: pediatric gastroenterologist, ped. nurse practitioner, behavioral psychologist, OT, SLP, feeding technicians, registered dietician, diet tech, nurses, licensed clin. social worker, case manger. 5 day/wk program, 6 hrs qd, including 4 therapeutic meals each day. A meal included oral motor exercises followed by oral feeding (for max of 20 minutes). All sessions videotaped and reviewed by caregivers and therapists. Weekly team meetings to review data, calories consumed, adjustments to tube feedings and tx/interventions for the coming week. Beckman oral motor exercises 20-30 min. prior to each oral feeding. (to increase functional response to pressure, movement). Meals completed by same staff members throughout the admission. Behavioral interventions incl. differential attn and prolonged presentation of food near the child’s lips until the child opened and accepted the food. Positive reinforcement. The therapists were the primary feeders until approx. 2 wks prior to d/c when caregivers were transitioned into meals. Parents were trained in food prep, calorie boosting and observed tx sessions. Caregivers move ahead when they had met >80% accuracy in instructions, prompts, and consequences training. When caregivers fed the child, the therapist observed and instructed.
RESULTS:
Results were reported in terms of statistical significance?
___ Yes
__x_ No
___ NA
___ Not addressed / What were the results? The authors do not explain their choice of analysis methods.
Outcomes / Results / Statistical Significance
Þ Meal-time behaviors (admission to d/c)
1. accept food by mouth
2. Mouth clean
3. Inappropriate behaviors
4. Duration of meal
5. Grams per meal
6. Calories consumed
7. Percent tube fed
8. Instruction, prompts, consequences (IPC)
Þ Weight
Þ Average % of expected weight gain
Þ Weight percentile (on growth curve)
Þ Mean percent ideal body weight
Þ Height
Þ Average % of expected height gain
Þ Height percentile (on growth curve) / 1. Improvement in ability to open mouth for food.
2. Gains in timeliness of managing food in mouth for more efficient chewing and swallowing.
3. Decreased inappropriate behaviors such as: refusing, vomiting, gagging, expelling, or holding food in mouth.
4. Tolerated longer meal sessions
5. Increased volume
6. Calories increased
7. Reduction in amt. of supplementation (due to increased oral consumption)
8. Improved ability of caregivers to feed children by providing appropriate IPC during meals.
Þ Improved admission to d/c, 1-4 months, 4-7 months, 7-12 months (0.2 kg “decrease” from d/c to 1 month)
Þ “Approached” weight gain expectations: 1-4 months, 4-7 months, 7-12 months (60-70% with very large SDs)
Þ Met or exceeded weight gain expectations: Admission to d/c, d/c-1 month, and d/c-12 months (again with large SDs).
Þ Improved at all intervals except at 1 month f/u. Improvement from below 3rd percentile to 10th percentile on growth curve from admission to 12 months (large SDs).
Þ Improved at all points except at 1 month post d/c.
Þ Improved at all points except at 1 month f/u (0.2 cm “decrease” d/c-1 month).
Þ “Approached” height gain expectations: 1-4 months and 7-12 months
Þ Met or exceeded height gain expectations: admission-d/c, d/c-1 month, 4-7 months, and d/c-12 months.
Þ Improved at d/c, 7, and 12 months. Improvement from 7th percentile to 16th percentile from admission to 12 months (large SDs). / 1. Not reported
2. “significant gains”- do we assume p-value <.05??
3. Not reported (mean decreased by more than ½, but SD remained relatively large)
4. Not reported
5. Not reported
6. Not reported (mean almost doubled, but SD is very large)
7. “significant reduction”- do we assume p-value <0.05??
8. Not reported (mean increased significantly)
Þ Not reported
Þ Not reported
Þ Not reported
Þ Not reported
Þ Not reported
Þ Not reported
Þ Not reported
Was the analysis, that is the type of statistically tests used, appropriate for the type of outcome measures and the methodology?
__X_ Yes
___ No
___ Not addressed / Explain: paired sample t-tests were used to look for significant changes in dependent variables, from adm to d/c and each follow up interval.
If not statistically significant (i.e., p < 0.05 or 0.01), was study big enough to show an important difference if it should occur (power and sample size)?
The sample size was certainly small and may not have provided enough power. It is difficult to know since statistical significance was not reported.
Clinical importance was reported?
___ Yes
__x_ No
___ Not addressed / What is the clinical importance of the results (that is even if the results were statistically significant were the differences large enough to be clinically meaningful?
Frequent/intense oral-motor exercises in combination with a behavioral approach can increase food consumption and decrease dependence on tube feedings. Due to unreported significance it is unknown if the resulting weight/height gain was more than just by chance.
Drop-outs were reported?
__x_ Yes
___ No / If yes, why did they drop out? How was drop-out participants included in the statistical analysis?
1 child was unavailable for follow-up due to hospitalization and distance for visits.
CONCLUSIONS AND CLINICAL IMPLECATIONS:
The conclusions made by the authors were appropriate given study methods and results.
__x Yes
___ No
Possibly improvements were seen, but whether this tx plan yields optimal feeding and growth is questionable. / What did the author concluded?
Improvement in oral intake can be accomplished through the combination of oral motor exercises, approp. body and head positioning, and positive reinforcement. Increased food consumption was a reflection of the child’s change in feeding skills and behaviors. Consumed calories improved. Caregivers improved their ability to feed their children, improvements in wgt gain improved for the entire group. The children were able to maintain their new weight percentile curve for the year following d/c.
What were the main limitations of the study as stated by the author(s) and from your point of view?
The authors did not discuss limitations. Limitations include: subject numbers, incomplete data analysis, parental/home/ environmental influences; Beckman Program may not be the optimal approach to use, the intervention and parent training strategies were unclear. Therapist involvement was also unclear. Who did the exercises and who fed the children is unclear – one therapist could have done it differently than another and it doesn’t seem that this was measured/evaluated. Different feeders could be using different approaches with each child, affecting each child’s outcome differently and potentially the entire group. Since the exercises were done prior to feeding there maybe a fatigue/attn/ behavioral factor here that wasn’t considered. The time was not specific and no explanation was offered. It is unclear whether the subjects received any other tx. at the same time. Nothing to this effect was stated. Given that it was a day tx. program, it is quite possible that they weren’t receiving other feeding-related tx, but it is unclear. It is also unclear whether the results could be related to maturation alone since results were not reported using statistical significance.
What are the implications of these results for your practice?
Validates the importance of parental involvement and need for a comprehensive approach, including biological and behavioral. The complexities of oral motor intervention requires inclusion of many factors- from providing optimal positioning to consistency/ continuity of the unique approach needed for individual success.

Potvin 2007 modified from Law, Stewart, Pollock, Letts, Bosch, & Westmorland, 1998