Feasibility of a Classification System for Physical Therapy, Occupational Therapy, and Sports Therapy Interventions for Mobility and Self-Care in Spinal Cord Injury Rehabilitation

Sample 

The study included all physical therapists, occupational therapists, and sports therapists (n=36) working with patients with SCI at 3 specialized Dutch rehabilitation centers. Therapists who were expected to be off-duty for more than half of the time during the study (eg, because they went on holiday) were excluded.

Treatment Selection 

All treatment sessions of patients with SCI at the inpatient and outpatient departments were included if (1) the patient was physically present, (2) the treatment session was an individual treatment, and (3) the treatment session was aimed at the domains of mobility and self-care.

Procedure 

After approval of the study by the scientific and ethical board of each center, the study was introduced during a special staff meeting, and therapists were asked to participate on a voluntary basis. All participating therapists (n=36) gave verbal consent to participate in the project. The therapists then received the classification system and a manual consisting of 3 parts: (1) general information on the structure of the classification system, (2) detailed descriptions of all coded interventions, and (3) a 1-page summary with coding instructions. The contents of and procedures for the use of the classification system were discussed with all therapists at a further instructional meeting. The therapists were then asked to practice recording their sessions with the classification system for a period of 1 week.

After this first week, a second instructional meeting was held at each center. Next, the participants were asked to record all relevant treatment sessions over a period of 2 to 4 weeks. After this period, the participants were sent a questionnaire on the feasibility of the classification system.

Measures 

To obtain data for evidence of completeness, the participants had to record their treatment sessions with codes. The form used to record treatment sessions allowed users to record a maximum of 5 different interventions a session and to indicate the amount of time spent on each intervention in 5-minute increments. If the therapists were unable to record an intervention in one of the listed categories, they could record it as unspecified. To obtain data for evidence of completeness, mutually exclusiveness, speed, and ease of use, each recording form also included 5 questions (questionnaire A) relating to (1) the difficulty users had when classifying interventions (7-point scale ranging from no doubt to too much doubt about which level, category, and intervention to choose), (2) (if therapists had been in doubt about which code to use) an open-ended question to describe what caused the doubt, (3) the time they needed to classify the activities in a treatment session (<1min, 1−3min, >3min), and (4) (if the manual had been consulted) the clarity of the general information in the manual, and (5) the detailed descriptions of interventions (7-point scale ranging from very clear to very unclear).

To obtain information related to the opinion of the therapists on completeness, mutually exclusiveness, speed, and ease of use, a questionnaire (questionnaire B) with 13 questions was administered at the end of the recording period. These questions related to the therapist's opinion on the instructional meeting, the various sections of the manual (7-point scale from very good to very poor), and whether they felt the average time needed to classify the activities in a treatment session was acceptable for research purposes and for daily use (7-point scale from very acceptable to very unacceptable).

Statistical Analysis 

Because the practice week did not reveal major problems, the data collected in this week were merged with the data collected in the other weeks. Descriptive statistics were used for the distribution of classification system codes and the time spent on each therapeutic activity. The percentage of unspecified codes was used as an indicator of the completeness of the classification system. Sufficient user satisfaction was defined as 80% or more of the therapists having a positive opinion on the feasibility of the classification system. To assess this, the 1 through 7 scales were dichotomized into positive opinions (scores 1–3) and neutral or negative opinions (scores 4–7).

Therapists 

A total of 20 physical therapists, 14 occupational therapists, and 2 sports therapists participated on a voluntary basis. The distribution of the 3 disciplines varied by center: in center A, 8 physical therapists and 5 occupational therapists participated; in center B, 8 physical therapists, 5 occupational therapists, and 1 sports therapist participated; and in center C, 4 physical therapists, 4 occupational therapists, and 1 sports therapist participated. One sports therapist of center A and 1 occupational therapist and physical therapist of center C were excluded according to the criteria. The participating physical therapists had been working with patients with SCI for a mean ± SD of 10.4±5.6 years (range, 3–20y) and the occupational therapists for a mean ± SD of 4.8±3.4 years (range, 1–10y). The sports therapists in centers B and C had been working with patients with SCI for 5 years and 1 year, respectively.

Treatment Characteristics 

The number of recorded treatment sessions varied in each center (257 in center A, 357 in center B, 242 in center C), by discipline (521 for physical therapy, 283 for occupational therapy, 52 for sports therapy), and by therapist (7–51). A total of 272 treatment sessions were recorded during the practice week, and 584 were recorded during the recording weeks. Treatment sessions of 142 different inpatients and outpatients with SCI were recorded. The number of patients did not differ significantly by center (45 in center A, 48 in center B, 49 in center C). The mean age of the patients was 49 years; they included 88 men and 54 women. The types of SCI included motor-complete tetraplegia (n=21), motor-complete paraplegia (n=26), motor-incomplete tetraplegia (n=42), and motor-incomplete paraplegia (n=53).

Codes and Time by Treatment Session 

The participants recorded a total of 1625 codes. Thirty-nine percent (335/856) of all treatment sessions required only 1 code to describe the therapeutic activities during the session, 32% (273) required 2 codes, 23% (199) required 3 codes, 5% (42) required 4 codes, and only 1% (7) required 5 codes. The mean time spent on the first coded activity varied from 15.8±8.9 minutes for physical therapists to 22.5±15 minutes for occupational therapists and 22.2±10.2 minutes for sports therapists. The largest amount of time spent on therapeutic activities by code was 60 minutes for physical therapists, 90 minutes for occupational therapists, and 45 minutes for the sports therapists.

Completeness 

The participants used interventions from all 3 levels. The physical and occupational therapists recorded at 3 levels of functioning, whereas the sports therapists recorded only at the level of basic functions and basic activities. Table 1 shows the numbers of codes in the various categories of the classification system. The therapists used 27 of the 28 categories. Only the category of bowel interventions at the level of complex activities was not used. Of the 27 categories used, 23 (85.1%) were used by at least 2 disciplines. The largest number of recordings occurred in the categories of muscle power, muscle length, walking, wheelchair driving, and handbiking. The physical therapists made most of their recordings in the categories of muscle power, muscle length, joint mobility, and walking. The occupational therapists made most of their recordings in the categories of wheelchair driving and handbiking, maintaining and changing body positions and movements, partial tasks of self-care, and arm and hand use. The largest numbers of recordings by the sports therapists were made in the categories of wheelchair driving and handbiking, swimming, arm and hand use, and muscle length.

Table 1. Distribution of the Recordings of Various Therapeutic Activities in Treatment Sessions in the Feasibility Study

	Category	Code	No.⁎	%†	Total Time‡
	Basic functions
	Muscle power	101	224	13.8	4220
	Muscle length	102	157	9.7	2755
	Muscle tone	103	80	4.9	1080
	Joint mobility	104	129	7.9	1910
	Sensory functions	105	7	0.4	125
	Neuropathic pain	106	5	0.3	45
	Musculoskeletal pain	107	23	1.4	445
	Skin and related structures (1)§	108	3	0.2	35
	Skin and related structures (2)§	109	42	2.6	965
	Cardiovascular system	110	34	2.1	655
	Respiratory system	111	20	1.2	210
	Total	NA	724	44.5	12,444
	Basic activities
	Hand and arm use	201	74	4.6	1215
	Subtasks of self-care	202	59	3.6	1270
	Basic body positions and movements	203	127	7.8	2110
	Transfers	204	84	5.2	1670
	Standing	205	89	5.5	1360
	Walking	206	150	9.2	2400
	Wheelchair driving and handbiking	207	140	8.6	3095
	Use of driving transportation	208	29	1.8	670
	Swimming	209	19	1.2	360
	Total	NA	771	47.5	14,150
	Complex activities
	Walking and moving around indoors	301	51	3.1	900
	Walking and moving around outside	302	24	1.5	465
	Transfers	303	20	1.2	240
	Washing and caring for body parts	304	12	0.7	260
	Toileting: bladder	305	1	0.1	45
	Toileting: bowel	306	0	NA	NA
	Dressing	307	16	1.0	500
	Eating and drinking	308	6	0.4	95
	Total	NA	130	8	2505
	Total of all recordings	NA	1625	100	29,099

Abbreviation: NA, not applicable.

⁎ Total number of recorded codes per category. † Percentage of the total number of recorded codes. ‡ Total time spent (in minutes) on therapeutic activities, per category. § There were 2 categories for skin and related structures: (1) general and (2) pressure ulcers.

All types of interventions, exercises and training, modalities, assessment, education, and equipment were used at all 3 levels. Of the 1625 codes, only 24 (1.9%) were recorded as unspecified. Another 42 codes (2.9%) were recorded for exercises and training without further specification.

Mutually Exclusiveness 

The therapists were able to classify the activities in most of the treatment sessions, with 767 (89.6%) of 856 involving little or no doubt. There was no significant difference in the level of doubt between the practice week and the recording weeks. A reason for doubt about the right classification was given in 140 cases. In only 27 of these 140 cases did the doubt concern the choice between the 3 levels of the classification system. Other comments concerned where to classify equipment, the difficulty in finding the category for the intervention (eg, walking the stairs, cycling), and difficulty in separating simultaneously performed interventions.

Speed 

The therapists recorded 756 (88.3%) of the 856 treatment sessions within 3 minutes and 388 (45.4%) of 856 within 1 minute. The time needed to record a treatment session was slightly longer in the practice week than in the recording weeks. The percentage of recordings taking more than 3 minutes was similar for the 2 periods (12.5% and 11.3%, respectively), but the percentage of recordings taking 1 to 3 minutes decreased (56.6% and 36.5%, respectively), while the percentage of recordings taking less than 1 minute was higher in the recording weeks (30.9%, 52.1%, respectively).

The therapists recorded most treatment sessions immediately after the session or later that same day. All therapists considered the time needed to classify a treatment session acceptable for recording for research purposes. Twenty-two of 36 therapists rated the time needed to classify a treatment session for daily use acceptable or neutral.

Ease of Use 

The therapists consulted the general information about the structure of the classification system in the manual in 251 (29.3%) of 856 of the recorded treatment sessions. If consulted, this information was rated by 193 (76.9%) of 251 as clear to very clear. The detailed descriptions of interventions in the manual were consulted in 405 (47.3%) of 856 of the recorded treatment sessions. If consulted, this information was rated by 363 (89.4%) of 405 as clear to very clear. Over 80% of the therapists had a favorable opinion on the general information, the detailed descriptions of the interventions, the recording instructions, and the recording form.

Comments 

The therapists were finally invited to mention possible improvements to the classification system. In most cases, the therapists suggested including more examples in the detailed descriptions of the interventions. The occupational therapists mentioned the difficulty in distinguishing self-care interventions at the level of basic activities from self-care interventions at the level of complex activities. The occupational therapists also mentioned the difficulty in classifying interventions involving equipment.

Study Limitations 

Although the data provide evidence for the feasibility of our recently developed classification system, this study was subject to certain limitations. Our results may not be generalized to other settings, because the classification system was developed for use in the Dutch SCI rehabilitation system, which might differ from that in other countries with respect to patient population and type of interventions applied. However, because we included a wide range of patients with SCI, from early admission to outpatient treatment, the sample might be considered representative of the whole spectrum of patients with SCI in postacute care. Further, international literature was used to identify relevant physical and occupational therapist interventions,18, 19 and our terminology was as much as possible derived from the ICF.14, 15

A second limitation is that the study included only 2 sports therapists, so we cannot conclude that the treatment sessions recorded in this study by the sports therapists were representative of the Dutch situation.

There were several comments from therapists that will be used to refine the classification system. Ongoing research includes more sports therapists and an investigation of interrater and intrarater reliability. Further research is currently examining the interrater and intrarater reliability of the classification system.