PROFILES OF OUTPATIENT SPEECH-LANGUAGE THERAPY AND AUDIOLOGY CLIENTS AT A TERTIARY HOSPITAL IN THE WESTERN CAPE

This study described trends in the demographic, diagnostic, and attendance variables o f first-tim e clients who attended the SpeechLanguage Therapy and Audiology outpatient department (OPD) at a tertiary hospital in the Western Cape over a five-year p er io d (19992003). A retrospective, descriptive survey was conducted and data were collected from hospital records. The data were coded, analysed descriptively and represented in tables, figures and graphs. The diagnostic categories o f hearing loss and voice disorder made up the highest percentage o f referrals. The largest percentages o f clients were referred from the Ear, Nose and Throat (ENT) Departm ent at the study hospital, fo llow ed by the education sector. Dem ographic trends indicated that the highest percentage o f clients attending the OPD was in the age category o f 7-12 years, that sim ilar percentages o f male and fem ale clients attended the OPD, and that English was the language that a client was most likely to speak. Geographically, the highest percentage o f clients attending the OPD was from the Klipfontein region. Attendance data revealed that approximately 70% o f clients attended OPD appointments within a three-month time p e ­ riod and attended no more than two appointments. Percentage o f appointments attended decreased with increasing numbers o f sched­ uled appointments. Implications fo r research and service delivery are discussed.


INTRODUCTION
This study aimed to describe the client population profiles of those attending the speech-language therapy and audiology outpatient department (OPD) at a tertiary hospital in the Western Cape in South Africa over a five year period (1999)(2000)(2001)(2002)(2003).To achieve this aim, the demographic, diagnostic and attendance data of a study sample were determined.The impetus for this study arose from difficulties the researcher experienced while attempt ing to engage with service planning for Speech-Language Ther apy services in a post-apartheid era at a tertiary hospital in the absence of knowledge of descriptive client data.
In the South African public sector, SLT & Audiology ser vices have traditionally been located in both the health and educa tion sectors.In the health sector these services have been consid ered as part of allied health service delivery with emphasis on rehabilitation (Allied Health Professionals Technical Committee, 2004).Although health services are essential, they fail to reach many people in South Africa due to a lack of resources.Histori cally, health service delivery in!South Africa has been shaped by the medical model and complied with apartheid ideology (Bhagwanjee & Stewart, 1999;Hall, Haynes & McCoy, 2002).Public health services were characterised by racial and geographi cal disparities, fragmentation and duplication of services and were hospital-centred with minimal emphasis on Primary Health Care (PHC).The combined influences of the medical model and apart heid ideology resulted in an inequitable provision o f services in terms of accessibility, appropriateness, funding and co-ordination across the variables of race, class, gender and level of urbanisa tion (Bhagwanjee & Stewart, 1999).The healthcare system was divided, inefficient and grossly inequitable (Hall et al., 2002).
Speech-Language Therapy (SLT) and Audiology Services in South Africa have been shaped by such socio-political contex tual realities and are also grossly inequitable (Pillay, Kathard & Samuel, 1997).Services in South Africa have been biased toward providing a better quality service to a White, middle class, Eng lish and Afrikaans first language speaking population, whilst pro viding a poorer service to a Black African first language speaking clientele (Pillay, 1996cited in Pillay et al., 1997).Public sector SLT & Audiology services have historically been based in hospi tals and special schools, with minimal service delivery at commu nity level.Similar to the national profile, services in the Western Cape have been urban-based with few Speech-Language Thera pists and Audiologists working in rural areas beyond the Cape Metropole.Health sector provision of SLT & Audiology services have been limited to services at tertiary hospitals, with little avail able at primary or secondary levels in the Western Cape (Allied Health Professionals Technical Committee, 2004), thus limiting access and availability o f services to the majority o f the popula tion.
The current service delivery situation is untenable given the prevalence of disability in the Western Cape.The data from the 2001 Census (Statistics South Africa, 2004) regarding the numbers of people with communication, hearing or multiple dis abilities point to the need for SLT & Audiology services in the City of Cape Town and in the Western Cape.The need for acces sible services, coupled with appropriate planning, and importantly the knowledge to assist such planning, is essential to meet the service delivery needs of this population.
Over the last decade, systemic changes have taken place within the health system which have had a bearing on SLT & Audiology services.Subsequent to South Africa's first democratic elections in 1994, the National Department of Health set about transforming and restructuring the healthcare system (Hall et al., 2002;Forman, Pillay & Sait, 2004).These changes included a policy shift to a primary health care (PHC) approach within a dis trict health system (DHS), with intent to realise a social model of health.The aim o f such transformation has been to improve the quality of care provided and create a more equitable service.There is a critical need for relevant knowledge to inform such planning processes.In order to ensure cost-effective improvement and development of existing health services, the efficacy of health programmes needs to be validated through research showing that such services are necessary, appropriate and accessible, within The South African Journal o f Communication Disorders, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)specified contexts (Bhagwanjee & Stewart, 1999).
Several studies within the SLT & Audiology profession, have examined client profiles to facilitate service planning.Inter nationally studies by Broomfield and Dodd (2004); Edwards, Cape and Brown (1989); Enderby and Davies (1989); Enderby and Petheram (2000); Heron (2001);and Petheram and Enderby (2001) have provided critical data to influence service planning.In South Africa a limited number o f studies have examined client data in different contexts (Klop, 1998;Schneider, 1992 andSwanepoel, 2005).Klop (1998) conducted a study in the area of quality man agement in a private healthcare practice in Cape Town.The out comes of the study were intended to help implement quality man agement programmes.Client profiles in terms of disorders, age, geographical location, gender, home language and referral agents were compiled from 197 clients seen by the researcher in her prac tice from 1994 to 1996.She found that the majority of clients at tended for language disorders, followed by stuttering, almost half were between three and six years of age, 63% of the clients were nale, almost three quarters of her clients were Afrikaans-speaking and most of the clients were drawn from an area within a 10 km radius o f the practice.Her clients were referred from other health care professionals, educators and former clients.From a smaller group of clients she found that only two out of 64 clients withdrew from therapy.Schneider (1992) collected data from records, reports and case files to determine the nature and prevalence of communica tion disorders seen in six hospitals in Gazankulu.Communitybased speech and hearing workers were then interviewed about their work situation, organisation of their time and intervention strategies used with communicatively disordered people in order to evaluate the efficacy of their work.The most common disorders seen at the hospitals in Gazankulu by these speech and hearing workers were hearing disorders.Swanepoel (2005) conducted an exploratory descriptive study which critically described an infant hearing screening pro gramme conducted at two maternal and child health clinics in Hammanskraal.Quantitative and qualitative methods for data col-!lection were used His study showed poor follow-up return rates i to the clinic.

|
In South Africa, apart from these studies by Klop (1998), f Schneider (1992) and Swanepoel (2005) there is limited data to contribute to service planning.The unique characteristics and challenges of developing countries demand that contextual and local research be done in conjunction with international .studies(Swanepoel, 2005).Speech-Language Therapy research in the United Kingdom has emphasised the importance of ongoing gathering information on services and client populations for Speech-Language Therapy services.As with South Africa, inequalities in services for the speech and language impaired in the United Kingdom's national health service have been documented (Enderby & Davies, 1989).If there is to be efficient planning in healthcare it is important that this process be informed with regard to the size and needs of the population to be served (Enderby & Davies, 1989).The collation of client data profiles may serve in questioning some of the basic premises regarding provision of services and should lead to con sideration of alternative forms of provision which meet the needs of the speech and/or language disabled and hearing impaired popu lation (Enderby, 1989).
To monitor equity concerns in South Africa, comparative data are necessary to understand the multifaceted nature of client profiles which include race, age, gender, urban/rural location and socio-economic status (Ntuli & Day, 2004).The first steps how ever, are to collect data from current service sites, as was the inten tion of the present study.The demographic, diagnostic and atten dance data of clients receiving services at a tertiary level o f health care could be valuable in planning equitable and appropriate ser vices.
Reliable data concerning the types of speech, language and hearing disorders referred, age and sources o f referral, and the ef fect of cultural and socio-economic profiles of the population on referral patterns are vital for planning services (Broomfield & Dodd, 2004).Comparison of records may show service trends, admission patterns and familial trends (Lubker & Tomblin, 1998) over time.It was important to collect data gathered over time so that changes in service provision could be examined with the pur pose of informing debate and planning (Petheram & Enderby, 2001).The compilation and analysis of data, timeous reporting and use of consistent up-to-date health information, are all key aspects o f healthcare planning and management.The lack o f information regarding the health sector has contributed significantly to the slow process of transforming the health system (Ijumba & Day, 2004).Thus the results of this study, with its focus on demographic, diag nostic and attendance data, has potential to influence service deliv ery changes with regard to SLT & Audiology services.

METHODOLOGY Aim
This study aimed to describe trends in the demographic, diagnostic, and attendance variables of first-time clients who at tended the SLT & Audiology outpatient department (OPD) at the tertiary hospital being studied over a five year time period (1999)(2000)(2001)(2002)(2003).

Study design
A retrospective descriptive survey was conducted, to collect demographic, diagnostic and attendance data of a sample of clients.

Sample i
The study population included all the hospital records of new clients who had attended the speech-language therapy and audiology OPD of the tertiary hospital under study during a 5 year period from January 1, 1999 to December 31, 2003.This period of time was selected as data were readily available and would give a sufficiently large number o f records for trends over the years to be examined.Due to the large number of clients in the population (2,819), a sample of one-third of the population data were drawn, using a stratified random sampling procedure (Katzenellenbogen, Joubert & Abdool Karim, 1997), stratified according to the year in which the client was first seen.The sample consisted of 929 cli ents, on which the analysis was conducted.

Data collection and analysis
The clients' names were retrieved from the client registers and record cards.The data collection form, devised by the re searcher, was then utilized to capture the data from the hospital records (Appendix A).Data were electronically captured onto a spreadsheet as raw data and then entries were coded by the re searcher.In instances where data were missing on the form (where it was not indicated as missing in the records), the records were Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)reviewed again by the researcher to retrieve this information.After the data had been entered into the computer and checked, client names and folder numbers were removed to ensure anonymity of the clients.Attendance data were then converted to number of appointments attended, time frame over which these appointments were attended and the percentage o f appointments attended.The coded data were transferred from the spreadsheet into a computerdriven statistical package (Statistical Package for the Social Sci ences) (SPSS Inc, 1995) and frequency and percentage distribu tions as well as contingency tables were generated.
In instances where a data collection form had missing data, after being re-checked by the researcher, information which was unavailable, was categorised as such.If a record for a client had missing data for two or three of the three diagnostic variables, the record was excluded from analysis of diagnostic variables and at tendance variables, but was still included in the analysis of demo graphic variables.In instances where there was no clear atten dance data for two or more attendance variables, that client was excluded from the analysis of attendance information.However, if only one of these variables was missing, the data were still in cluded in the overall attendance analysis.
Due to the descriptive nature of this study, analysis of variables was conducted via descriptive statistics (Hite, 2001), as used in other similar studies (Enderby & Petheram, 2000;Petheram & Enderby, 2001).
Frequency and percentage distributions were used to describe the demographic, diagnostic and attendance data.
Frequency data were converted into percentage data for each year and for the total sample.Contingency tables (cross tabulation) were used to show relationships between variables.Inferential statistics were not used as this form of data analysis is not appropriate for a descriptive study (Hite, 2001).

Reliability and validity
Reliability of the data collection procedure was ensured by defining the exact methods of measuring and checking of work (Katzenellenbogen, et al., 1997).In this study a standard, reliable data collection form (Appendix A) was used to collect data.Cross checks o f information from the hospital record sheet to the SLT & Audiology folder also enhanced J the reliability of the data collec tion process.| Areas o f concern for validity included the reality that hospital records are generally produced for clinical, administrative or monetary ends rather than research purposes (Abramson, 1990) and that different clinicians ma>| have used different terminology when describing diagnoses.Terminologies were therefore exam ined and then grouped in broad categories.Data collection was verified by the researcher as it was entered into the computer.
Instrument, observer and client variations can be evaluated by repeating measures o f a sub-sample of the study sample (Katzenellenbogen, et al., 1997).In this study a random sample of ten percent of the clients from each year was selected for re collection of data (n=93) by an independent researcher.The data were then coded and the codes were compared to those found for the same clients by the researcher.There was 95% agreement across all the variables, indicating good inter-observer reliability.
A large sample was chosen to enhance the validity of the study as confidence intervals become narrower as sample size in creases (Katzenellenbogen, et al., 1997).The data were analysed quantitatively using SPSS to ensure the data were analysed in a consistent manner thereby ensuring valid and reliable results.The results were interpreted by the researcher who also engaged with a process of peer review to enhance the quality of the interpretation.

Ethical considerations
The head o f the speech-language therapy and audiology department, the medical superintendent of the tertiary hospital be ing studied, and the ethics committee of the University of Cape Town were consulted with regard to gaining access to the hospital files.It was not, however, possible to gain consent from the cli ents, due to potential difficulties locating clients and the number of records being reviewed.However, confidentiality was strictly maintained by removing any identifying information after the data had been checked and coded.Access to the files may be consid ered ethical on the grounds that there was no risk of harm to indi viduals, there was potential public benefit, and investigators' pro tected the confidentiality of the individuals whose data they stud ied (Council for International Organisations of Medical Sciences, 1991).The community will stand to benefit from this research if changes, taking into account the findings of this study, are made within the healthcare setting with regard to equity in service provi sion and aligning the services to the needs of the population.

RESULTS AND DISCUSSION
The number of clients varied each year ranging from 178 to 197 clients, with a total sample size of 929 clients on which these results are based.The results and discussion are presented in rela tion to the aims of the study.The total number of clients varied across the aims because of missing information as explained in the methodology.The quantitative results are represented in tables and graphs and the main trends are described and then discussed.

Speech-Language Therapy and Audiology Diagnosis
; In this study recorded reported medical diagnoses have been managed as inter-related diagnostic categories with SLT & Audiology diagnoses.The percentage of clients within each SLT & Audiology diagnostic category is presented in Figure 1.j Speech-Language Therapy and Audiology diagnostic categories of j those attending the OPD appeared to remain fairly consistent over the five-year period, with minor variations across the years.

! j
The common diagnoses of those attending the SLT & Audi ology OPD included hearing loss (n=255; 30%), voice disorders (n=l 77; 21%) and neurogenic communication disorders.The ; types and percentages of diagnoses remained relatively similar across the five years.
There are several significant issues surrounding the types of disorders seen at the SLT & Audiology OPD and the percentage of clients attending the OPD for the management of each disorder (Figure 1).The reason for the high percentages of clients attend ing the OPD as a result of a hearing loss or voice disorders, may be due to the fact that this SLT & Audiology OPD is situated in a tertiary hospital, with specialized equipment and facilities suitable for assessing and treating these disorders.
Due to the fact that hearing loss and voice disorders, as well as neurogenic communication disorders (for which the third highj est percentage of clients attended the OPD) are medically based and are treated by the medical profession in conjunction with Speech-Language Therapists and Audiologists, clients may be likely to attend this hospital which offers both services.Hearing loss, voice and neurogenic communication disorders are often ac quired and therefore there is a 'loss' of function, making the disor der noticeable and igniting a desire or need to restore lost function.The proximity and connections to the ENT department, who made the largest percent age of referrals to the SLT & Audiology OPD, may also account for the high per centages of clients who attended the OPD as a result of a hearing loss or voice disor der.
The highest percentages of clients were diagnosed with hearing losses, voice disorders, neurogenic communication dis orders or fluency disorders, which are 'noticeable' disorders and thus easier to detect than less 'visible' disorders (McLaren, Solarsh & Saloojee, 2004) and are likely to impact on everyday life.Less 'visible' disorders, such as a language im pairment, may not be as easily noticeable in everyday situations and thus not as eas ily detected.These less 'visible' disorders are then not as likely to be referred to SLT & Audiology services.It is also possible that because the Department is actually referred to as a "speech therapy" rather than "speech-language therapy", it may not be obviously associated with managing language disorders, a reality that the pro fession has experienced internationally.

Referral Sources
The results for referral sources ap pear in Figure 2. From the year 2000 on wards there was an increase in referrals from community clinics and from wards at the study hospital.
When examining individual cate gories of clients with a specific disorder or from a specific referral source, particu lar trends were found.These results are based on the clients within a specific cate gory only and thus the percentages and numbers are based on these individual categories and not on the full sample.The majority of clients who attended the SLT & Audiology OPD as a result of a voice disorders (n=137; 76%) were re ferred from the ENT department.Fifty percent (n=41) of referrals from commu nity clinics, 80.5% (n=33) of referrals from occupational health centres and 83% (n=10) of referrals from old age homes or residential care were referred to the SLT & Audiology OPD as a result of a hearing loss.The majority of the referrals to the OPD from wards were for dysphagia, neurogenic communication disorders or a combination of these (n=31; 97%).The education sector referred 50% (n=28) of paediatric clients with multiple difficulties to the OPD.
Referral sources (Figure 2) would also have shaped the client diagnostic pro files discussed above.As mentioned, the ENT department made a high percentage of referrals to the SLT & Audiology OPD.This trend is understandable in light of the discussion around the types of disorders seen most frequently at the OPD.The second major referral agent to the OPD was the Education Sector.This situation may be attributed to the limited SLT & Audiology services within mainstream schools (F.Lewis, personal communica tion, February 18, 2005), resulting in re ferrals to the Health sector.j I Demographic data | I

Age Groups
Age group results, presented in Table 1, illustrate that the largest percent age (n=l 65; 18%) of clients who attended the OPD were within the age group of 7 -12 years.
The following results are based on sub-categories of clients.Clients attend ing the OPD with a hearing loss or tinni tus covered the full range of ages, al though the majority of these clients were between 30 and 79 years of age (n= 181; 72%).The majority of clients who at tended the SLT & Audiology OPD as a result of a voice disorder were between 30 <9* %.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) ancj 49 years of age (n=98; 54%).Clients with a neurogenic com munication disorder spanned a wide range o f ages, although the highest percentage fell within the age group of 50-59 years (n=27; 24%).The majority o f clients who attended the OPD as a result of dysphagia were between 50 and 69 years of age (n=10; 67%), while clients with both dysphagia and a neurogenic communica tion disorder were slightly older, between 60 and 79 years (n=23; 62%).The majority of clients presenting with fluency disorders, however, were under 18 years of age (n=64; 90%).A similar trend occurred for clients attending as a result o f a phonology/ articulation/oral motor disorder, language disorder, language and learning disorder or multiple difficulties although the majority were under the age of 12 (n=20; 87%).While the highest percent age of clients attending as a result o f language disorders was in the under six year age group (n=15; 65%), children attending the SLT & Audiology OPD with language and learning disorders or multi ple difficulties had a greater percentage in the 7-12 year old cate gory (n=25; 83% and n=35; 62.5% respectively).
Despite the high number of clients with a hearing loss or voice disorder attending the OPD (which in this study were found predominantly in the adult population), the majority of the caseload attending the SLT & Audiology OPD was found to be between 7 and 12 years of age (Table 1).Of this age group only a minority attended the OPD for a hearing loss or voice disorder.However, this age group had the majority of the clients with flu ency disorders, phonology/articulation/oral motor disorders, lan guage and learning disorders, as well as those with multiple diffi culties.A reason for this high percentage of 7 to 12 year olds at tending the OPD under study in relation to the under six popula tion, may be due to the close proximity of another tertiary hospital, which specializes in paediatric intervention.The high percentage of 7 to 12 year olds attending the OPD (who are primary school age children) alludes to the lack of services within the education system, who were also the second highest referral agent.Despite the inclusive education policy intention regarding inclusive educa tion (Department of Education, 2001), there seems to have been minimal support for learners with communication difficulties in mainstream schools in the Western Cape.
The researcher has also perceived a general lack o f knowl- edge amongst educators, many health professionals and the public, of SLT & Audiology services, resulting in 'late' referrals of indi viduals.Anecdotal parent reports have indicated that parents who had voiced their concerns regarding their child's delayed language development have been told by health professionals 'not to worry' or that their child will 'grow out of it'.Thus, instead o f the child receiving early intervention, intervention was only available upon entering school where there are high demands on communication skills and the child's difficulties begin to impact on academic per formance.This scenario again relates to the 'visibility' of certain disorders (McLaren et al., 2004), as these language and learning disorders are often only detected in an academic environment and are not immediately obvious in everyday situations (Das, 2001).
Examination of the results pertaining to the adult popula tion attending the Speech-Language Therapy and Audiology OPD showed that there was a slight decrease in percentage of clients attending the OPD within each 10 year category from 30 to 69 years, with a slightly larger decrease in therapy attendance at 70 to79 years and a large decrease in therapy attendance in the over 80 year old category.An explanation for the greatest number of adults attending the OPD being between 30 and 49 years of age may be due to the large number of clients attended the OPD as a result of a voice disorder and the fact that the majority of clients with voice disorders were in this age group.
This finding o f a decrease in therapy attendance amongst clients with increasing age, however, seems to follow a general trend in rehabilitation, where services are less accessible and avail able to the older population.Although disability is more prevalent in the older population, it seems that older adults are unable to access hospital based outpatient services.Blake (1981) in the United States of America, claims that in terms of age group, par ticipation in the rehabilitation service seems to be inversely related to the need for such service.Rehabilitation may be more of a pri ority for younger people, or services may be more accessible to them, than for the older population.Rehabilitation has previously been driven both philosophically and financially by the goal of restoring individuals to productive employment, resulting in better opportunities for younger clients, although this is now changing to include a focus on older clients (Raia, 1992).The motivation to attend rehabilitation may be greater for the younger adults due to their need to regain employment, as well as to pursue social goals.A study at Groote Schuur Hospital in the Western Cape, South Africa (Whitelaw, Meyer, Bawa & Jennings, 1994) confirmed this trend and reported that greater numbers of stroke clients under the age of 65 years old, who had been inpatients, were referred (as outpatients) to Physiotherapy and Occupational Therapy services, while fewer clients above the age of 65 were referred.Further more, a greater number of the younger adults presented for and received therapy than older adults.Another study in the Western Cape, examining rehabilitation services at Bishop Lavis Rehabili tation Centre, documented that 50% of the population attending these services were 60 years old or younger (Rhoda, 2001).
In light of the above discussion regarding services for dif ferent age groups of clients, the current study revealed that a large percentage (n=71; 64%) of clients with a neurogenic communica tion disorder (without dysphagia) were under the age of 60 years.However, 70% (n=26) of those with both dysphagia and a neuro genic communication disorder were over the age of 60.This pos sibly indicates a greater degree of disability in the older age groups, where multiple disabilities are perhaps more likely.This trend of multiple disabilities in the older population may also ex plain the reduction in the number of these older clients attending outpatient rehabilitation.Elderly clients may not only be more The South African Journal o f Communication Disorders, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)affected by a stroke, but may also have more problems reaching therapists than younger clients (Whitelaw et al., 1994).There is also greater attrition of older clients than younger clients due to stroke (Whitelaw et al., 1994).

Gender
Slightly more than half of the sample who attended the OPD was male (n=470; 52%).More females than males presented with voice disorders, language disorders or no specific Speech-Language &/or Audiology diagnosis.
It was unexpected to find similar percentages of male and female clients attending the OPD, as the belief was that there would be many more males than females as found in other studies (for ex ample Broomfield & Dodd, 2004;Enderby & Petheram, 2000;Klop, 1998), although direct comparisons could not be drawn as some studies did not include Audiology clients and some were based only on paediatric populations.This trend, however, be comes clearer when one considers gender in relation to the different disorders.
As found in the present study, as well as other studies (Coyle, Weinrich & Stemple, 2001), voice disorders are more prevalent amongst females than males.This study found that a large percentage of clients (n=180; 21%) attended the OPD for the management of a voice disorder and because many of these clients were female the number of females attending the SLT & Audiology OPD was increased.
There are two possible explanations as to why there were more female than male language disorders found in this study.Firstly, more males presented with language disorders coupled with learning disorders and were thus categorised as such.Secondly, it was found that more of the language disorders fell into the 0-6 year old category.There is a perception that 'boys develop more slowly' or 'boys are slower to talk' (Phillips, 2004:500) and they might therefore be brought later for therapy, when they would more likely be diagnosed as having a language and learning difficulty.
Of all the other disorders (apart from language disorders) found mostly in children, all had higher percentages of males than females.Many studies have commented on the greater likelihood of reading difficulties or dyslexia in males than in females (Owens, 1999) and that children with dyslexia have delayed language devel opment.Also, as expected, more males than females attended the OPD for the intervention of fluency disorders (Guitar, 1998).

Home Language
The majority of clients attending the OPD were classified as having English or Afrikaans as their home language, with a minor ity speaking Xhosa or another language (Table 2).

Table 2: Number and percentage of clients per Language group who attended the Speech-Language Therapy and Audiology OPD at a tertiary hospital in the Western Cape from 1999-2003 (n=910)
This is interesting considering that 55% of the population in the Western Cape speak Afrikaans as a home language, 24% speak Xhosa as a home language and 19% speak English as a home lan guage (South Africa.info reporter, 2002).There are several possi ble reasons for the mismatch between the population profile and the profile of clients attending the SLT & Audiology OPD.
Firstly, English may be the preferred language of interven tion for several reasons.The majority of Speech-Language Thera pists and Audiologists in the SLT & Audiology OPD at this tertiary hospital in the Western Cape were English first language speaking with Afrikaans as a second language.Also, many children were being schooled in English, as English was the language most com monly used as the medium of instruction (Alant, 1989).Thus they may choose to receive therapy in English.Thirdly, there were no Xhosa-speaking Speech-Language Therapists or Audiologists at the study hospital and no formal access to interpreters, resulting in lim ited provision of services to the Xhosa-peaking population.This situation is of grave concern given the multilingual nature of the population in the region.
Another possible reason for Xhosa-speaking clients not ac cessing services in the OPD clinic may be due to cultural reasons.Cultural groups vary in their view of disability and therapeutic in terventions and such differences may impact who is likely to attend the hospital (Swartz, 1998;Swanepoel, 2005).Furthermore, the hospital in the present study has historically been a 'white' hospital, and there may still be misconceptions about who can access ser vices at this hospital.In many communities in South Africa there is little awareness of SLT & Audiology services, due to lack of re sources and facilities, and therefore people are unaware that ser vices are available.
Access to services is also influenced by socioeconomic status of people.Although poverty is not confined to one racial group in South Africa, it is most prevalent amongst the black popu lation (Swanepoel, 2005).Many of the Xhosa-speaking population in the Western Cape, the majority of whom would be black, may have a limited income.The costs of travel to the hospital as well as multiple treatment fees may be prohibitive.

Geographical Location
The Health Districts in Cape Town were used to categorise the areas from where the clients who attended the OPDlcame.i Table 3 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) Table 3 shows the number and percentage of clients from each area.It can be seen that clients attended from many different areas.This is not sur prising given the lack of SLT & Audiol ogy services in the community or even at a secondary level of care (Allied Health Professionals Technical Commit tee, 2004).Some of the areas, however, would have been covered by SLT & Audiology services at another tertiary hospital in Cape Town.

Attendance
The majority of clients attended the OPD over a very short period of time, for example many attended within a three month period (n=513; 71%) or attended two appointments (n=506; 70%).Number of appointments and time frame results showed very similar trends, with a drop-off with increasing number of scheduled appointments or longer period of time over which these appointments were scheduled.Almost half of all the clients (n=353; 49%) at tended only one appointment.Only 9% (n=67) of clients attended more than five appointments.
Given that long-term therapy was often indicated, these find ings were noteworthy in relation to ser vice planning.
The high number of clients who only had one booked appointment may be due to the fact that certain clients would have been placed on a waiting list for therapy and therefore not been given further scheduled appointments immedi ately.They may then not have wanted therapy when it was offered, due to changes in circumstances between the time of the assessment and the date when therapy was offered.Additionally, there are some cases for which one attendance may be all that is indicated.Some clients may have needed an initial assessment with no necessary follow-up.It is likely that once their concerns had been ad dressed, they no longer felt the need for further intervention.Enderby and Da vies (1989) reported that 60% of newly referred children required assessment and advice only.While this explanation might be applicable to some cases, the data in case files, the nature of the disor ders and personal experience suggest that many required additional intervention.
The lack of understanding of the nature of interventions offered by the profession of SLT & Audiology by the community, as well as the location of the services within a tertiary hospital envi ronment, may have contributed to poor attendance rates.Many clients have the expectation that a hospital is a place which one attends when ill and that it is a once-off attendance, possibly with a fol low-up appointment.Many expect treat ment that is tangible, such as medication or surgery.Clients arrive with the ex pectation that they will be given some thing to 'cure' them in a once-off ap pointment.Therefore, they may not have been prepared for the active role that is required of them in the therapeutic proc ess.Also, they may have been unwilling or unable to commit themselves to fur ther intervention as recommended.The high rate of non-attendance may also reflect the economically disadvantaged nature of the population served, and the low priority given to SLT & Audiology services given other life circumstances (Broomfield & Dodd, 2004).
A further explanation for low at tendance rates could relate to different perceptions of a disorder by a therapist and a client, or the client having a differ ent understanding of the nature of a dis order.Negotiating between explana tions of illness increases the possibilities of compliance or adherence to treatment (Swartz, 1998).Thus, explaining the nature of the diagnosis in ways that the client can understand and relate to their understanding of the world, may im prove attendance and compliance to therapy.Also, discussions prior to com mencing therapy about the nature of therapy and the respective roles of the parent and the therapist may well in crease parent satisfaction (Roulstone, Glogowska, Peters & Enderby, 2004), which may improve attendance rates for children.The same principle could be applied to adults.
With an increasing number of scheduled appointments, percentage of appointments attended dropped.The only diagnostic categories which had more than 10% of clients attending in excess of five appointments were for the categories of fluency disorders (n=15; 28%), paediatrics with multiple difficul ties (n=12; 25%), language disorders (n=5; 24%) and phonology/ articulation/ oral motor disorders (n=3; 13%).It is possible that the motivation for treat ment of these disorders may have been greater than for other disorders.
Data presented by Enderby and Davies (1989), which included many but not all types of communication disorders included in this study, indicated that the children who received regular therapy, attended once a week for an average of 16 weeks.Roulstone et al. (2004) found that of the 71 children allocated to Speech-Language Therapy in their study, 68 attended sessions offered, al though nearly 18% of appointments were either cancelled (by the therapist or clinician) or not attended.Klop (1998) in her private practice found that only two out of 64 clients dropped out of therapy.The differences in findings be tween these studies confirm the impor tance of understanding how contextual realities influence service provision.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)

LIMITATIONS
There are several limitations of this study which need to be considered.There has been minimal research into client profiles attending SLT & Audiology services in South Africa.Therefore, the researcher had to rely on the methodology and literature from other countries and other healthcare disciplines to conduct this study.While such literature was useful, it was not directly appli cable to the SLT & Audiology contexts in South Africa.The study was also limited to the data from one tertiary hospital examined over a five year time period.The conclusions therefore are only applicable to the tertiary hospital being studied during the speci fied time period.
Due to the nature of the study, records used were not de vised specifically for research purposes and thus may not have been maintained with the care that would be expected in a planned investigation.Notably, as diagnostic labels vary in use across SLT & Audiology services and medical categories, specific diagnoses are open to different interpretations.For example, a patient's pri mary medical diagnoses may be a "cerebral-vascular accident" with SLT & Audiology services referring to "neurogenic acquired communication disorders".This resulted in difficulties during the data collection phase which included missing records, incomplete records, and information recorded in an inconsistent manner within and between records.This is a common limitation of retrospective studies such as those by Farmer (1990) and Schneider (1992).A further problem associated with the description of communication disorders is that many disorders (for example learning difficulties) are not clearly defined (Lubker, 1997).
Limitations in the interpretation of the results included lack of comparison data from other sites, limited ability to generalize the findings of this research to other service sites, and that the in terpretation was speculative in nature.Due to the descriptive na ture of the study, no causal or correlational assertions could be drawn (Hite, 2001).Despite such limitations, the findings can in form the restructuring of services in the region.
The difficulties encountered with the use of records points to the need for good record keeping, which is essential for plan ning of healthcare services (Rhoda, 2001).Inadequate records can distort research results, prevent clients from being involved in re search and negate the quality of research (Farmer, 1990).There fore, methods of recording client information and treatments used need to be reviewed (Farmer, 1990).This information is not only useful for research, but also when assessing cost-efficiency of a service (Rhoda, 2001).
The limited access to SLT & Audiology services in the Western Cape, as reflected in the poor attendance rates obtained in this study, point to the need for SLT & Audiology services at all levels, and particularly at a primary level, of care (Rhoda, 2001).Services need to be provided in line with the Primary Health Care Approach within the District Health System (Hall et al., 2002;For man, et al., 2004).Ideally, the services should be integrated at all levels of care and should strive towards community based rehabili tation.As a starting point, therapists' participation in community outreach programmes would be advantageous.

CONCLUSION
This study found that the highest percentages of clients at tending the SLT & Audiology OPD at a tertiary hospital in the Western Cape attended for a hearing loss (n=250; 29%) or voice disorder (N=180; 21%) and that the ENT department at the study hospital (n=203; 24%) and the education sector (n=l 11; 13%) re ferred the highest percentages of clients.In terms of age, the highest percentage of clients attending the OPD were between 7 and 12 years old (n=165; 18%).Overall there were similar percentages of male and female clients who attended the OPD.English speaking clients predominated (n=501; 55%), followed by Afrikaans speak ing clients (n=298; 33%).Only a small percentage of clients were Xhosa speaking (n=82; 9%).Patterns of attendance showed that almost half of all the clients who attended the OPD attended only one appointment (n=353; 49%).In total just over 70% of clients attended either no more than two appointments (n=506; 70%) or attended within a three month period (n=513; 71%).With an in creasing number of scheduled appointments, percentage of ap pointments attended decreased.
Given the poor attendance rates and the high number of cli ents in the 7 to 12 year age group seen at the SLT & Audiology OPD, there is a definite need for more adequate service provision in schools.The Department of Education in the Western Cape has very limited SLT & Audiology services (F.Lewis, personal com munication, February 18, 2005).However, the Department of Education (2001) emphasises the need to optimize the expertise of specialist support personnel, such as therapists, psychologists, re medial educators and health professionals within the school set ting.In the long term therefore, the service provision for learners in inclusive settings must be given urgent consideration.Cur rently, the Education sector in the Western Cape does not have the capacity for individual intervention in ordinary classrooms, but plans to provide preventative services and services to educators (F.Lewis, personal communication, February 18, 2005).
One of the major challenges facing the SLT & Audiology profession is to provide equitable services for the multilingual populations.The scarcity of Xhosa-speaking therapists in the Western Cape is a source of concern.In the long term it would be advantageous to train and employ professionals speaking African languages.However, given the language diversity in the country, it is critical that practitioners become skilled to manage multilin gual and multicultural populations (Swanepoel, 2005).For exam ple, the inclusion of interpreters to support clinical practice is im portant.However, there were no available interpreters at the SLT & Audiology OPD under study, despite repeated requests to the hospital authorities in this regard.While service delivery to multi lingual populations remains a challenge, it is important that steps are taken to create equitable practices.
The South African Journal o f Communication Disorders,Vol.53, 2006    Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)V S p e e c h a n d A u d i o l o g y D ia g n o s is

Figure 1 :
Figure 1: Percentage of clients per Speech-Language Therapy and Audiology diagnostic category who attended the Speech-Language Therapy and Audiology OPD at a tertiary hospital in the Western Cape from 1999-2003 (n=859).

Figure 2 :
Figure 2: Percentage of clients per referral source who attended the Speech-Language Therapy and Audiology OPD at a tertiary hospital in the Western Cape from 1999-2003 (n=859).Abbreviations used: C T : Cape Town OAH: Old Age Homes ENT: Ear, Nose, Throat

Figure 3 :
Figure 3: Percentage of clients who attended their scheduled appointments at the Speech-Language Therapy and Audiology OPD at a tertiary hospital in the Western Cape from 1999-2003 across different time periods (n=719).

: Number and percentage of clients per Geographical location who attended the Speech-Language Therapy and Audiology OPD at a tertiary hospital in the Western Cape from 1999-2003 (n=910)
Die Suid-Afrikaanse Tydskrifvir Kommunikasieafwykings,Vol. 53, 2006