Home health physical therapy: practice patterns in western New York.

Home health care services are increasing in the

United States. Data from the 1993 National

Home and Hospice Care Survey show that

there was a 33% increase in the number of

Medicare-certified home health care agencies between

1989 and 1994.[1] Three primary factors contributed to

the rapid growth of these agencies: (1) for many

patients, home health care can be provided at a lower

cost than care in an institution,[1-3] (2) Medicare

regulations provide reimbursement for home services to

elderly persons, and (3) the explosive growth in the

segment of the population aged 65 years and older,

which has increased the demand for home care services.[1]

The 1995 membership profile of the American Physical

Therapy Association shows that increasing numbers of

therapists work in home health care.[4] Physical therapy

ranks third, behind nursing and personal care services,

as the most frequently used type of home care service.[1]

In 1993, 82% of patients who received home health care

needed skilled nursing care, 59% required personal

care, and 18% used physical therapy services.[1] Although

physical therapy is provided to a large number of

patients, there is little information regarding the

practice patterns of physical therapists in the home,

including the types of interventions provided and how often

these interventions are administered. Although home

care is not a new idea, it is now being delivered by more

providers to diverse patients.[5]

A recent study conducted in the United Kingdom

examined the distribution of work time for community physical

therapists.[6] The data obtained from a survey of time

allocation in a 1-week period indicate that 29% to 45% of

the therapists' time is spent in face-to-face encounters with

patients and that 23% to 30% of the therapists' time is

devoted to patient support activities such as consultation

with other providers, case conferences, ordering of

equipment, documenting work, and scheduling appointments.

The remaining time was spent in travel (19%-21%),

administration (11%-13%), studying (1.7%-2.7%), and

training other staff (0.4%-3.2%).[6]

Although the British study provides some insight into

the general distribution of work time of community

therapists, the types of activities and the distribution of

time spent in home care may differ in the United States

for several reasons. The therapists who participated in

the British study were employed full-time as community

physical therapists.[6] In the United States, home health

physical therapy is often provided by therapists who work

only part-time in this area.[4] This difference may make a

difference in how time is spent in home care. In

addition, the policies for payment of home health services

differ between the United States and the United

Kingdom. In a survey of community therapists in the Trent

region of the United Kingdom, Seymour and Kerr[7]

found that about 66% of therapists reported no

restrictions on the number of visits allowed for their patients.

In the United States, third-party payers often place limits

on the number of visits that will be reimbursed. Finally,

the mode of practice of community physical therapy in

the United Kingdom differs from home health physical

therapy in the United States. Community physical

therapists provide the majority of their treatments in

patients' homes, but they also provide services in satellite

clinics and community hospitals.[7] In the United States,

home health care is usually considered distinct from

services provided in clinical settings. This difference in

practice patterns may have an impact on the types of

treatment provided and on the distribution of work

time.

Information about home health practice is important for

the physical therapy profession from several

perspectives. If the profession is to be responsive to the

increased utilization of physical therapy in home care,

academic programs need to prepare graduates to

function effectively in that environment. This preparation

should be based on the characteristics of actual practice

in the home. Experienced practitioners may also benefit

from new information and continuing education in

home care physical therapy. We believe that it is the

profession's responsibility to investigate the effectiveness

of physical therapy provided in the home through

outcome studies. We contend, however, that a basic

understanding of what happens in the home is necessary

before outcomes can be studied.

The purpose of our study was to identify the types of

patient care activities performed by home health

physical therapists and the time spent in these activities. A

secondary purpose was to determine

whether the time spent in different

home health activities varied according

to patient age or diagnosis. Although

many people may perceive home care

as predominantly serving older adults,

only about 59% of home care service

fees are paid by Medicare, and only

50.5% of home care visits are made to

individuals 65 years of age and older.[8]

Variation by patient age or diagnosis

would indicate that these factors may

be important in planning therapists'

caseloads or patient programs. Factors

such as these are important for case

managers to better understand the role

of the physical therapist in patient

management and for rehabilitation managers

to consider in allocating staff or

determining the case mix of a particular

therapist.

Method

Instruments

A log sheet for recording home health

activities was developed based on our

home health experiences and with

input from other physical therapists

and therapist managers who worked in

home health. This instrument was pilot

tested with five physical therapists, who

completed the log sheet for 1 week.

The pilot test provided feedback

regarding the categories and types of

activities that were included on the log

sheet, the format of the recording

sheet, and the time required to complete the log sheet.

Based on this feedback, the instrument was revised to

include additional physical therapy activities that were

not included on the initial log sheet. The final format

included 50 activities engaged in by the physical

therapists in 10 general categories: evaluation, treatment,

education, environmental management, supervision,

communication, consultation, documentation,

administration, and travel/other. The Figure shows the activities

in each of the general categories. Therapists were

instructed to count an activity only once. For example,

although documentation could be considered by some

to be an activity in the administration category on our

log sheet, it should have been recorded only once in the

column for documentation. Therapists were provided

with operational definitions for each category (Figure).

Finally, therapists were instructed to contact the

researcher if any questions arose once they began using

the instrument.

Figure.

Log sheet activities.

Evaluation

Initial

Vital Signs

Reevaluation

Discharge

Other

Treatment

Exercise

Functional Training

Transfer Training

Gait Training

Modalities

Orthoses/Prostheses

Balance Exercise

Endurance Training

Massage

Other

Education (any teaching regarding the

patient's program)

Patient

Family

Home Health Aide

Physical Therapist Assistant

Other

Environmental Management

Evaluation

Modification

Order Equipment

Adjust Equipment

Other

Supervison (monitoring of patient

care-related activities provided by others)

Physical Therapist Assistant

Home Health Aide

Family

Student

Other

Communication (with patient/family

other than teaching)

Introduction

Explanation of Services

Schedule

Other

Consultation (with other health care

professionals)

Physician

Nurse

Occupational Therapist

Speech/Language Therapist

Physical Therapist

Other

Documentation

Progress Notes

Open Case

Close Case/Discharge

Other

Administration

Department Meeting

Team Meeting

In-service

Other

Travel/Other

Travel

Other

Subjects

Physical therapists who worked in home health care were

recruited from three home health agencies in

metropolitan Rochester, NY. Thirty-two therapists initially

expressed an interest in participating and were trained

in the data collection procedures. In these training

sessions, operational definitions (as shown in the Figure)

were provided and participants' questions regarding

completion of forms were answered. Of these initial

subjects, 19 therapists gave informed consent to

participate and submitted completed data forms. Eighteen

participants were female, and 1 therapist did not

complete the gender item on the demographic

questionnaire. Characteristics of the participants are summarized

in Table 1.

[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]

Procedure

Over a 2-week period the therapists recorded the

number of minutes they spent on the activities associated

with their home care responsibilities. The data were

recorded by patient visit; by time spent performing

duties on behalf of a patient, even when not performed

at the patient's home; and by time spent performing

other activities not associated with a specific patient (eg,

meetings). Therapists also completed a patient

information sheet, recording the age and primary diagnosis of

each patient treated during the 2-week period. The

reliability and validity of the responses on the log sheet

are not known, so the results related to these factors

should be considered exploratory.

Data Analysis

Frequencies of interventions and the means and

standard deviations for time spent on each task were

calculated using DataDesk 4.0 for the Macintosh.(*) Data were

analyzed for all patients together, as well as according to

patient age (0-20, 21-64, and 65+ years) and patient

diagnosis. The patient diagnoses were grouped into four

categories: neurological (eg, Parkinson's disease,

cerebrovascular accident), orthopedic (eg, fractured hip,

total joint replacement), cardiopulmonary (eg,

congestive heart failure, emphysema), and other (eg, dementia,

frequent falls, open wounds). For all analyses, the

average amount of time spent in interventions was calculated

based on the number of visits that included that

intervention. For example, vital signs were monitored in only

38 of the visits. Average time spent monitoring vital signs

was calculated by averaging the time spent on this

evaluation activity during those 38 visits, rather than by

averaging the time spent in this activity across all 737

visits recorded for this study. One-way analyses of

variance (ANOVAs) were used to determine differences in

treatment based on patient age and diagnostic category.

An ANOVA was performed for each group of therapist

activities as well as for the total time spent per visit.

Tukey's Honestly Significant Difference test was used for

post hoc testing. Differences were considered significant

at the .05 level of probability. Although the number of

statistical tests performed resulted in a relatively high

cumulative alpha level, we believed that this level was

appropriate because the study was exploratory in nature.

Results

Therapists recorded activities associated with 737 visits

for 279 patients. Four hundred sixty-three visits (62.8%)

were made to patients aged 65 years or older, 222 visits

(30.1 %) were made to patients aged 21 to 64 years, and

52 visits (7.1%) were made to patients aged 20 years or

younger. The average age of patients treated by home

health physical therapists was 64.3 years (SD = 21.8). Two

hundred twenty-seven visits were associated with patients

with neurologic diagnoses, 352 visits were associated with

patients with orthopedic diagnoses, 59 visits were

associated with patients with cardiopulmonary diagnoses, and

99 visits were associated with patients with "other"

diagnoses. Table 2 shows the breakdown of patients by age

group and diagnostic category.

Table 2.

Characteristics of Patients Receiving Home Health Care

                                     Percentage of

Age Group/Diagnosis                  Therapists(a)              N



0-20 years (n=15)

  Neurologic                             3.2                    9

  Orthopedic                             0.4                    1

  Cardiopulmonary                        0.4                    1

  Other                                  1.4                    4



21-64 years (n=82)

  Neurologic                            10.0                   28

  Orthopedic                            15.1                   42

  Cardiopulmonary                        1.1                    3

  Other                                  3.2                    9



65 + years (n=182)

  Neurologic                            16.5                   46

  Orthopedic                            28.7                   80

  Cardiopulmonary                       11.5                   32

  Other                                  8.6                   24

(a) Percentages are based on total of 279 patients.

The frequency of visits for general categories of activities

and the average time spent in each category are shown in

Table 3. The most frequently performed activities

associated with home care were in the categories of

documentation (performed in 91% of the recorded visits),

travel/"other" (performed in 90% of the recorded

visits), and treatment (performed in 89% of the recorded

visits). The majority of time in home care was spent in

direct patient treatment as opposed to other associated

activities such as education or evaluation. When visits

included treatments, therapists spent an average of 35

minutes on specific treatment activities. Administrative

duties were also time consuming but were performed

less frequently than treatments. In the 34 visits that

included administrative tasks, therapists spent an

average of 22 minutes on these duties.

[TABULAR DATA 3 NOT REPRODUCIBLE IN ASCII]

Travel and documentation in the form of progress notes

were the most frequently performed activities associated

with home care, followed by exercise, gait training, and

patient education (Tab. 4). Although travel and

documentation were performed more frequently, time spent

in treatment was more than double the time spent in

either of these categories for all age and diagnostic

groups. The most time-consuming activity,

"administrative other," was performed in only five visits but required

an average of 44 minutes over those five occurrences.

Activities included in this category were specialized

meetings or care conferences. Initial evaluations were

also time consuming. In the 75 visits in which initial

evaluations were performed, an average of 29 minutes

was spent doing these evaluations.

[TABULAR DATA 4 NOT REPRODUCIBLE IN ASCII]

Table 5 shows the data analyzed according to the patient

age group. Documentation, travel/"other," and

treatment were the three most common categories of

activities associated with home health visits, regardless of

patient age group. Average time spent in these general

categories of activities varied according to patient age

group. Younger patients (0-21 years of age) received

longer treatments than patients in the other age groups

received, and they required more travel time than older

patients (65+ years of age) required. Older patients

required more documentation time than younger

patients (0-21 years of age) required.

[TABULAR DATA 5 NOT REPRODUCIBLE IN ASCII]

The average time reported as being spent in general

activities by patient diagnosis is shown in Table 6.

Treatment was the most time-consuming activity for all

four diagnostic groups. The ANOVAs showed

differences among diagnostic groups in the average amount

of time spent in patient treatment, environmental

management, consultation, travel, and total visit time (Tab.

6). Patients with neurologic diagnoses received longer

treatments than patients in any of the other diagnostic

groups received. Patients with cardiopulmonary, and

orthopedic diagnoses received less time in

environmental management than did patients with neurologic and

"other" diagnoses. Patients with neurological diagnoses

received more time in consultation than did patients in

any other diagnostic group. It took therapists longer to

travel to patients with "other" diagnoses than it took to

travel to patients with orthopedic or cardiopulmonary

diagnoses. Patients with neurologic diagnoses required

more time per visit than did patients with orthopedic or

"other" problems.

[TABULAR DATA 6 NOT REPRODUCIBLE IN ASCII]

Discussion

This study provides insight into the multiple roles of

home health physical therapists. Not surprisingly, the

study showed that the majority of home health

therapists' time is spent in treatment activities. The most

common types of treatments performed in the homes

also are not surprising. Although we did not examine the

functional limitations of the patients, given the

limitations that would be expected for home bound

individuals, gait training and exercise could be expected to be

the most widely used treatments.

The results indicate that physical therapists spend

substantial amounts of time performing activities other than

treatment. Fifty-four percent of all visits that were

recorded for this study included some time in education,

with an average of 14 minutes per instance. In a recent

study of patient education behaviors in physical therapy

practice, Gahimer and Domholdt[9] recorded the number

and type of educational statements that occurred during

outpatient sessions. The greatest number of educational

statements occurred in the categories of information

about illness and home exercises.[9] Although we did not

examine the types of education statements made, the

considerable time spent providing patient education

serves to further substantiate Gahimer and Domholdt's

point regarding the importance of the physical

therapist's role as an educator. In-service training regarding

effective teaching in the home may be helpful to better

prepare physical therapy practitioners to fulfill the

educator role in this setting. Likewise, these data suggest

that academic programs need to emphasize patient and

family education as an important skill for physical

therapists. Because we examined time spent on activities, our

suggestions are based on the assumption that therapists

should be competent in activities that take the most

time, not on whether these activities led to beneficial

outcomes.

Therapists working in home care do not have the

advantage of close proximity to other therapists or

health care providers that therapists working in clinics

may enjoy. This characteristic of home care may lead to

the assumption that home health physical therapists

work in isolation. The therapists in this study did not

participate in any formal team meetings during the 2

weeks in which the data were collected. Therapists

participating in this study, however, consulted with other

team members in 29% of the visits recorded. This

finding indicates that the therapists sought out the

expertise of other professionals as part of the ongoing

treatment process. We believe, therefore, that home

health therapists should be able to interact effectively,

with other members of the health care team. The team

approach allows for more comprehensive overall care.[10]

The provision of such care requires that each team

member apply his or her professional knowledge and

skills while taking other team members' contributions

into account. This approach requires not only technical

competence but also skills in communication, negotiation,

and conflict resolution.[10] The data from our study

suggest that these skills need to be emphasized in both

academic and clinical preparation of physical therapists

and in continuing education for the practicing therapist.

The safety of patients' home environments is a critical

factor in the patients' ability to function independently.

This area has been highlighted in recent literature,

especially in regard to the incidence of falls among older

people.[11-13] Approximately one third of the older

population falls each year, and one third to one half of those

falls are attributed to environmental hazards.[14] Home

health therapists are in an ideal situation to enhance

safety by identifying hazards and modifying

environments. Approximately 20% of the visits recorded in our

study included some form of environmental

management. Because home health therapists have ready access

to the patient's home environment, they can potentially

have a large impact on the safety of the environment.

This is an area in which therapists working in home care

may require training to enhance their competence in

identifying and modifying safety hazards. Several

instruments have been developed to identify environmental

hazards in the home.15,16 Competence in home hazard

identification could help physical therapists train other

team members and family members to recognize and

modify environmental hazards.

The differences between the amount of time reported as

being spent in treatment and patient age were of interest

to us. Although the total amount of time per visit did not

differ according to patient age, older patients received

shorter treatments than younger patients received.

These findings are consistent with those of Rohan and

colleagues' study of oncological social work.[17] Those

researchers found that social workers at one facility

provided fewer individual sessions, counseled on illness

adaptation less frequently, and spent less time with older

patients than with younger patients. Rohan et al

suggested that ageist biases contribute to these differences

in care. Similar studies of service utilization in hospitals

indicate that age is a factor in determining the amount

and type of medical care provided, with older patients

receiving less care than younger patients.[18-20] The data

from our study indicate that similar disparities in the

provision of care may exist in home health physical

therapy. In physical therapy, there are many reasons that

patients may receive shorter treatments. Older patients

may be unable to tolerate longer sessions due to the

severity of their illness, and treatments may be

abbreviated to prevent fatigue. Alternatively, therapists may

treat older patients for shorter periods because of lower

expectations for improvement. We did not control for

illness severity or the multiple factors that may influence

how time is spent in a home health visit. Future research

in this area should examine these factors.

In our study, visits to patients with neurological

problems took longer than visits to patients with other

diagnoses. Patients with. neurological diagnoses received

longer treatments, more time for environmental

management, and more time in consultation with other

health care practitioners. These findings may indicate

that patients with neurological diagnoses have complex

problems that consume more time during home visits.

The complexity and the severity of patient problems

were not addressed in our study; therefore, our

explanation is based on conjecture.

Less time was spent addressing environmental

management issues during visits to patients with orthopedic or

cardiopulmonary diagnoses than during visits to patients

with neurological problems. Environmental management

included ordering and adjusting equipment and

making recommendations for home modification.

These findings suggest that patients with neurological

diagnoses require more adjustment to their home

environment, but a comparison of the amounts of time spent

in environmental management recorded at intervals

over the entire course of therapy would provide more

information on whether these patients actually require

more of these modifications.

We found that therapists spend an average of 74 minutes

per home health visit. This finding was interesting

because the majority of the therapists participating in

this study (63%) were paid by patient visit rather than on

the basis of salary. This payment method might seem to

some to encourage short visits, but we found that

therapists spent relatively long periods of time working

with their patients during home health visits. Because we

used self-reports in our study, however, the possibility

exists that time reports were influenced by the

therapists' beliefs of what amounts of time might be

considered professionally acceptable. The low number of

therapists providing data for this study precluded analysis of

differences in treatment times based on therapist

characteristics. Whether differences existed in the amount of

time spent per visit based on payment method,

therefore, could not be determined.

Lang[6] conducted a similar study of time allocation in a

region of the United Kingdom. Although the therapists

in our study recorded time allocated by patient visit,

Lang recorded time spent over the course of a typical

week. In our study, average time spent in activities was

calculated based on the number of visits that included

that activity, so the actual percentage of time spent in

activities per visit cannot be directly calculated from the

data reported. Lang found that therapists spent

approximately one third of their time in activities not directly

related to patient care, such as travel and administrative

work.[6] The therapists in our study also reported

spending a relatively large amount of time in activities not

directly related to patient care, such as travel and

documentation. Due to differences in the categories

used for recording data, other comparisons regarding

the specific types of activities performed by therapists

cannot be made.

Several limitations of this study should be noted. The

first limitation is that the study was based on therapists'

self-reports of how they spend their time. Self-report

measures are known to be susceptible to bias,

particularly if the topic is perceived to involve

self-incriminating behaviors.[21] Although steps were taken to reassure

participants that results would be seen only by the

researchers and not shared with employers, there may have been

a tendency for participants to complete the log sheet in

a manner that reflects positively on the way in which

their work time is spent.

A second limitation is that short-term data collection was

used, with data recorded for patient visit activities over

only 2 weeks. Because many patients in home care may

receive physical therapy over a longer period, the data

may not accurately reflect the relative proportions of

time spent in each activity area. For example, patient

education may occur early in the course of home health

physical therapy or toward the end of a course of therapy

in preparation for discharge. If a patient whose data

were included in this study was in the middle of a course

of therapy, she or he may not have required education,

so no time would have been recorded for this activity for

that particular patient. Similarly, environmental

modification may be required in only one or two visits over the

course of physical therapy. These activities could have

been performed during periods in which data were not

collected, thus decreasing the apparent average time or

frequency of this type of intervention. Data collected

over the entire episode of patient care (from physical

therapy initial evaluation to discharge) would provide a

more complete and accurate picture of how home

health physical therapists work with patients.

Another limitation is that only physical therapy services

were examined. No note was made of what other services

were being provided to a patient. In many cases, there

may be potential for overlap between the services

provided to a patient, particularly in the case of a patient

receiving both occupational therapy and physical

therapy. If a patient was receiving both of these services, it is

likely that much of the functional training that a physical

therapist might otherwise perform (eg, tub transfer

training, practice in reaching items on high shelves or

on the floor) would be performed by the occupational

therapist. We did not collect data on the other services

provided to the patients in this study, so it is impossible

to determine whether the mix of services received by a

patient influenced the type of treatments provided by a

physical therapist.

The small number of therapists providing data for this

study is a limitation that precludes analysis based on

therapist characteristics. Therapists' educational

background, area of expertise, or experience in home health

could affect the amount of time spent in the various

activities associated with home care. In addition, the

study was conducted in one metropolitan area in western

New York. Future research with a larger sample of

therapists in diverse regions of the country might

provide important information about how therapist and

geographical characteristics affect the way home health

physical therapy is delivered.

Perhaps another factor that could influence the type of

physical therapist interventions is the patient's status at

the initiation of treatment. Because home health services

are generally very functionally oriented, the patient's

functional status is probably a very important

determinant of the type of interventions provided by a physical

therapist in the home setting. Likewise, a patient's living

situation may affect the types of home care

interventions. Patients who live alone may need to attain very

different levels of functional independence compared

with patients who live with a spouse or other caregiver.

Thus, the types of services provided by the physical

therapist may differ substantially for patients with

different living situations. Other factors that may affect physical

therapist interventions include the patient's

premorbid lifestyle, gender, marital status, and comorbid

conditions.

Examination of the outcomes of home health physical

therapy, especially as they relate to the functional status

of patients who are homebound, is needed. Currently,

there is a paucity of data on the effectiveness of home

care rehabilitation services. In one study,[22] a specialized

occupational therapy intervention for patients with

rheumatoid arthritis was associated with improved

functional status of the participants. Another study[23]

demonstrated improvement in functional measures of 16

patients receiving home physical therapy. Given the

large percentage of patients who receive physical

therapy in home health care settings, it seems prudent to

examine how physical therapy affects function and the

long-term use of health care services.

Summary

Our study examined the self-reported practice patterns

of home health physical therapists in upstate New York

who indicated the relative amount of time spent in 50

activities associated with home care. Therapists

participating in this study reported spending an average of 74

minutes per patient visit, with most of the time spent in

direct patient care. Over half of the visits recorded for

this study included some form of patient, family, or staff

education, indicating that home health physical

therapists should be competent teachers. Home health

physical therapists also consult with other health care

providers and therefore should be able to communicate

effectively with other members of the health care team.

Older patients received shorter treatments than young

patients received, but they required longer times for

documentation. Patients with neurologic diagnoses

received longer treatments, more environmental

management, and more consultation than patients with

other diagnoses received, perhaps reflecting the

complexity of the problems experienced by patients with

these disorders. Further research is needed to examine

how the variance in treatment time and mix of

treatments affects the outcomes of home health physical

therapy.

References

[1] Dey AN. Characteristics of Elderly Home Health Care Users: Data from the

1993 National Home and Hospice Care Survey -- advance Data From Vital

and Health Statistics, No. 272. Hyattsville, Md: National Center for

Health Statistics; 1996.

[2] Harrow BS, Tennstedt SL, McKinlay JB. How costly is it to care for

disabled elders in a community setting? Gerontologist. 1995;35:803-813.

[3] Braun KL, Rose CL. Geriatric patient outcomes and costs in three

settings: nursing home, foster family, and own home. J Am Geriatr Soc.

1987;35:387-397.

[4] 1995 Active Member Profile Report. Alexandria, Va: American Physical

Therapy Association; 1995.

[5] May BJ. Home Health and Rehabilitation. Philadelphia, Pa: FA Davis Co;

1993.

[6] Lang C. A survey of how community physiotherapists use their time.

Physiotherapy. 1996;82:222-226.

[7] Seymour JS, Kerr KM. Community-based physiotherapy in the Trent

region: a survey. Physiotherapy. 1996;82:514-520.

[8] Altman B, Walden D. Home Health Care: Use, Expenditures, and Sources

of Payment -- National Medical Expenditure Survey Research Findings 15.

Rockville, Md: Agency for Health Care Policy and Research, Public

Health Service; 1993. AHCPR Publication No. 93-0040.

[9] Gahimer JE, Domholdt E. Amount of patient education in physical

therapy practice and perceived effects. Phys Ther. 1996;76:1089-1096.

[10] Purtilo R. Health Professional and Patient Information. Philadelphia,

Pa: WB Saunders Co; 1990.

[11] Campbell A, Borrie MJ, Spears GF. Risk factors for falls in a

community-based prospective study of people 70 years and older.

J Gerontol. 1989;44:112-117.

[12] Tideiksaar R. Falls among the elderly: a community prevention

program. Am J Public Health. 1992;82:892- 893.

[13] Northridge ME, Nevitt MC, Kelsey JL, Link B. Home hazards and

falls in the elderly: the role of health and functional status. Am J Public

Health. 1995;85:509-514.

[14] Rubenstein LS, Robbins AS, Schulman BL, et al. Falls and instability

in the elderly. J Am Geriatr Soc. 1988;36:266-278.

[15] Nevitt MC, Cummings SR, Hudes ES. Risk factors for injurious falls:

a prospective study. J Gerontol. 1991;46:M164-M170.

[16] Rodriquez JG, Baughman AL, Sattin RW, et al. A standardized

instrument to assess hazards for falls in the home of older persons.

Accid Anal Prev. 1995;27:625-631.

[17] Rohan FA, Berkman B, Walker S, Holmes W. The geriatric oncology

patient: ageism in social work practice. Journal of Gerontology and Social

Work. 1994;23:201-221.

[18] Rosenthal GE, Fortinsky RH. Differences in the treatment of

patients with acute myocardial infarction according to age. J AM Geriatr

Soc. 1994;42:826-832.

[19] Hoenig J, Rubenstein L, Kahn K Rehabilitation after hip fracture:

Equal opportunity for all? Arch Phys Med Rehabil. 1996;77:58-63,

[20] Hamel MB, Phillips RS, Teno JM, ct al. Seriously, ill hospitalized

adults: Do we spend less on older patients? J Am Geriatr Soc. 1996;44:

1043-1048.

[21] Oyster CK, Harten WP, Llovens LA. Introduction to Research.

Philadelphia, Pa: JB Lippincott Co; 1987.

[22] Helewa A, Goldsmith CH, Lee P, et al. Effects of occupational

therapy home service on patients with rheumatoid arthritis. Lancel.

1991;337:1453-4456.

[23] Bohannon RW. Physical performance measures for the elderly

rehabilitated at home. Int J Rehabil Res. 1997;20:107-109.