Severe disability : Do primary health care nurses have a role to play ?

This article focuses on the importance of primary health care nurses’ involvement in the identification of chil­ dren with severe disabilities, early and appropriate inter­ vention that includes referral, as well as the provision of support to the children’s caregivers. The use of multi­ skilling as a strategy to train nurses to fulfil this role is described. The traditional roles of community nurses are explored within the disability paradigm, with specific reference to multi-skilling. Finally, research results fol­ lowing the implementation of the Beginning Communi­ cation Intervention Protocol (BCIP), which uses multi­ skilling as a training strategy, are described. Recommen­ dations for further research are then provided. Opsomming H ierdie artikel fokus op die belangrikheid van gemeenskapsverpleegkundiges se betrokkenheid by die identifikasie, vroeë en toepaslike verwysing en intervensie van kinders met erge gestremdheid en die ondersteuning van hulle versorgers. Die gebruik van veelvuldige vaardighede as ‘n strateg ic om verpleegkundiges op te lei om hierdie rol te vervul word bespreek. Die tradisionele rolle van gemeenskaps­ verpleegkundiges word in die gestremdheidsparadigma ondersoek, met spesifieke verwysing na veelvuldige vaardighede. Ten slotte word navorsingsbevindinge bespreek na die implementering van die Beginnende Kommunikasie Intervensie Protokol (BKIP), wat veelvuldige vaardighede as opleidingsstrategie gebruik. Aanbevelings vir verdere navorsing word dan verskaf.


Introduction
In an earlier article Alant (1998:20) stressed the need for primary health care nurses to become more involved in the identification of and intervention with children with severe disabilities (CSDs), in particular those who have little or no functional speech.This need for nursing intervention is based on the fact that the incidence of non-speech is no ticeably higher in South African schools for children with intellectual impairments than in schools reflected in com parable studies from other countries.For example, the South African figure was 38% (Bornman & Alant, 1997:17) as op posed to comparable data from North Dakota, USA which reported a rate of 2,4% (Burd, Hammes, Bomhoeft & Fisher, 1988:371) and in rural areas of Washington State, USA where 6% was reported (Matas, Mathy-Laikko, Beukelman & Legresley, 1985:20).There are a number of reasons for this high incidence.One of the most prominent is the lack of appropriate rehabilitation due to the limited number of trained professionals (e.g.therapists, teachers, nurses) in this field.In addition, services are often inaccessible as the majority of them are located in a few large cities, and trans port difficulties in rural areas are notorious.
Despite this bleak picture, community health clinics that should render comprehensive integrated primary health care (PHC) services using a one-stop approach, are accessible to the majority of South Africans (Department of Health, 1999:13).Primary health care nurses are often the first pro fessionals who come into contact with children with severe disabilities and their caregivers and/or parents (Clark, 1996:61).With the move towards the inclusion of people with severe disabilities, the majority of these individuals live with their parents (or extended families) and thus need to adapt to community living, making it mandatory that their needs be viewed within the context of the family and the community (National Department of Education, 2002:20).
Primary health care nurses are often the only professionals who provide continual support and assistance to these These specific nursing tasks and roles at each multi-skilling level, as well their appli cation to disability, are depicted in Figure 1 (American Speech and Hearing Associa tion 1996:56, Hurst, 1999:170;Pietranton & Lynch, 1995:38, South African Nursing Council, 1985:3;South African Nursing C ouncil, 1984:2;W ilkey & Gardner, 1999:303).
All primary health care nurses have the role of nursing practitioner.Community health nursing includes both personal health (fo cused on the individual and mostly includ ing the maintenance of health and recov ery from illness) and public health (promot ing and protecting the health of the com munity) (Thomas, 1999:17).These domains relate to the first level of multi-skilling, namely cross-training in basic patient care skills.Because the activities at this level mostly refer to traditional nursing tasks covered by the basic nursing curriculum (South African Nursing Council, 1985:2), they will not be further discussed.individuals and their caregivers, because nurses are read ily accessible and often act as the bridge between other professionals and the caregivers.Furthermore, as nurses are trained in active listening, caregivers perceive them as approachable and concerned advisors who can identify with the problems that concern the family, and who will help the family achieve their best possible health condition (Hitchcock, 1999:426).In view of the shortage of qualified health care professionals in South Africa (Bortz, Jardine & Tshule, 1996:467), primary health care nurses are ideally positioned to provide services to caregivers of CSDs (Moodley, Louw & Hugo, 2000:26).They are equipped to perform this task because their training provides them with the skills to observe behavioural patterns and environmen tal concerns, which enables them to make recommenda tions where necessary.It is therefore clear that primary health care nurses need to be equipped with the necessary knowl edge and skills to function as part of a health care team when assisting caregivers in dealing with their CSD's.
One way in which primary health care nurses can be equipped to do this, is though multi-skilling.Multi-skilled health professionals can be defined as persons crosstrained to provide more than one function, often in more than one discipline.These combined functions can be found in a broad spectrum o f health-related jobs, ranging in complexity from the non-professional to the professional level, including both clinical and management functions.Traditional functions (skills) added to the original health care worker' s job may be o f a higher, lower or parallel At the second multi-skilling level, the pri mary health care nurse has three major nurs ing roles: providing effective patient advocacy, executing education programmes and co-ordinating the health care regimens provided for the individual (South African Nurs ing Council, 1984:1).The advocate role requires the pri mary health care nurse to speak or act on behalf of indi viduals who may be unable to speak for themselves (e.g.CSDs and their caregivers) as a result of limited knowl edge; difficulty or inability in articulating their own needs or ideas; fear; perceived lack of power; and/or intellectual or physical disability (Wilkey & Gardner, 1999:306).Advo cacy also entails the preparation of individuals to stand alone and to speak for themselves rather than remaining dependent on the nurse (e.g. by introducing them to disa bled people's organisations) (Clark, 1996:64).As a result of their close contact with CSDs and their caregivers, nurses are often the best health professionals to promote the needs and desires of these children.One of the major nursing tasks would be raising awareness about disability issues in the community; for example, by giving health talks.
Linked closely to the advocate role is that of educator.Health teaching and the provision of information are viewed as essential nursing responsibilities.Education can be con ducted at the individual level (assisting CSDs and caregivers to make informed decisions about rehabilitation) or at a community level (e.g.health talks about disability) (Wilkey & Gardner, 1999:309).
Thirdly, the nurses act as co-ordinators or case managers of the health care regimens at the second level of multi skilling.This role implies the application of strategies to co-ordinate and allocate services for individuals who can not m anage th eir own care (e.g.CSD s and their disempowered caregivers) (Lemer & Ross, 1991:47;Wilkey & Gardner, 1999:312).The primary nursing tasks as applied to CSDs, and their caregivers would entail identification of CSDs and referral when needed.These tasks highlight the importance of feedback and follow-up in the attempt to monitor a child's progress.Primary health care nurses should be sensitive to the needs and circumstances of the caregivers, because they may appear to be non-compliant (e.g. may not attend referral and/or not return for feedback and follow-up) if they do not share the professional's val ues and priorities.This failure in turn leads the nurse to experience frustration and hostility (Humphry, 1995:692).
At the third multi-skilling level, primary health care nurses act primarily as educators and researchers.The importance of education has already been highlighted.At the third level, education is specifically related to the prevention of disability (at primary, secondary and tertiary levels).En hancing this preventative function evinces an enquiring and scientific approach requiring primary health care nurses to also act as researcher (South African Nursing Council, 1985:3;Wilkey & Gardner, 1999:310).This may include tasks such as identifying problem areas; collecting, analysing and interpreting data; applying findings; and evaluating, designing and conducting research (Griffith, 1994:69;Wilkey & Gardner, 1999:310).This thinking requires sensitivity to wards evidence-based practice.
Finally, at the fourth multi-skilling level, the nurses have two main functions -those of clinician and of collaborator.The role of clinician at this level is broader than on the first level, because it also entails an intervention function, al beit limited.At this level of multi-skilling nurses act as col laborative professionals moving the strict traditional boundaries of their discipline to facilitate the planning of further services, namely identification, screening and plan ning of services for CSDs.Full case management itself would thus entail counselling on prevention, follow-up, providing medication, collaboration and referral (Wilkey & Gardner, 1999:302).The collaborator role is closely linked to the provision of services, and also to other roles (e.g.advocate and educator).Consultancy requires that primary health care nurses help individuals to understand their dis abilities and make informed decisions about their own re habilitation (American Association of Colleges of Nursing, 1996:119;Wilkey & Gardner, 1999:307).
When looking at the four multi-skilling levels, it is impor tant to note that these skills cannot be viewed as distinct categories, and that extensive overlapping occurs.The lev els are also dynamic in nature and may change over time, or according to the needs of the CSDs and their caregivers.These needs are in turn influenced by, among other fac tors, the specific disability type.However, in an attempt to clarify specific nursing tasks as they pertain to disability, these categorical distinctions were made.Questions also remain unanswered regarding the application of multi skilling within various health care settings (e.g. would multi skilling be applied differently in primary, secondary and tertiary health care settings?),geographical areas (e.g.ru ral vs. urban), clinical disciplines and professions (e.g.nurs ing or speech and language pathology) (Pietranton & Lynch, 1995:38) and national policies and frameworks (e.g.how will the National Qualification Framework that endorses learning and skills acquired through experience and in-serv ice training, view multi-skilling?)(Geyer, 1997:11).Another grave concern about multi-skilling is that it might add to the workload of already over-burdened primary health care nurses.Professionals working in the health care arena are diverse regarding education, experience, training, autonomy and level of client contact, and therefore a unified approach to multi-skilling seems unlikely.
The current article will argue that primary health care nurses should be equipped, through multi-skilling, to function as part of a multi-disciplinary team, capable of delivering serv ices to one of the most neglected groups of individuals in their clinics, viz.those with severe disabilities.

Research results on multi skilling
The aim of the research was to train nurses in applying the Beginning Communication Intervention (BCIP) protocol aimed at equipping them with basic knowledge and skills related to service provision to CSD's and their caregivers.This protocol is aimed at multi-skilling levels 2,3 and 4 (See figure 1).This research formed part of a PhD study; twenty primary health care nurses from the Moretele Health Dis trict (in the North West Province) were trained using multi skilling principles (Bomman, 2001:4-25).Research consisted of a pre-experimental and an experimental phase.

Pre-experimental phase
The first phase of the research was to conduct a needs analysis by means of focus group discussions (qualitative data) and a questionnaire (quantitative data) in order to (i) ascertain the perceptions of community nurses regarding service delivery to CSDs, (ii) determine nurses' exposure to CSDs, and (iii) establish how primary health care nurses perceived the needs of CSDs and their caregivers.Results indicated that nurses did indeed feel that they had a role to play in providing services to this population, and that their current roles mostly included obtaining a case history, conducting a physical examination, detecting the problem, conducting health promotion (e.g. through health talks), treatment of minor medical ailments, referral and fol low-up (Bomman, 2001:6-6).However, they also expressed feelings of inadequacy because they did not know how to handle the caregivers of CSDs, depression, attitudinal bar riers (due to limited staff and time constraints) and the lack of follow-ups.In order to address feelings of inadequacy, the second phase of the research commenced.

Experimental phase Aims
The main aims of this phase were to describe the multi

Research design and phases
A quasi-experimental time series group design O^X-Oj-Oj-0 4-0 j with withdrawal was used as it was appropriate to determine knowledge and skills acquired over time (Brink, 1999:105).These phases all followed a linear course and are presented in figure 2.

Participant selection and description
Two selection criteria were used, namely that the nurses had to be employed by the Moretele Health District and that they had to be working in either a community health clinic or a mobile clinic.Non-probability sampling tech niques were used to select participants.Firstly purposeful sampling was used as the assistant Director of Health in the particular area selected nurses who were eager to re ceive further training in this field.Secondly a convenience sample was used in order to least disrupt service delivery (e.g.nurses from the same clinic were not used, and neither were nurses who were on leave, close to retirement, on night duty, etc.).Although these sampling techniques had many advantages for the present research, it hampered generalisability of results (Brink, 1999:140).This resulted in the selection of 20 primary health care nurses.Descriptive

Instruments used
In order to meet the require ments posed by the research question and design, two measuring instruments, re sponse forms I and II, were developed, as well as the BCIP training procedure.They can briefly be d e scribed as follows: Response form I This structured interview dealt primarily with the skills and applied knowledge of the nurses and was based on a specific case study.A dif ferent, but comparable case study was used for each of the research phases in order to counteract familiarity.The four different case studies that were used are presented in Appendix II.Response form I consisted of five sec tions, namely the biographic data, the applied knowledge about the current abilities of the particular case study, ap plied knowledge about recommendations, a practical skill demonstration in using the BCIP with the particular case and finally their exposure to the BCIP.An outline is pre sented in table 2 .
Basic considerations had to be taken into account when conducting these structured interviews and completing response form I as they impacted on the reliability and va lidity of the data.This included the following: All interviewers received the same training before the commencement of the structured interviews and the completion of response form I. In addition, an external rater, (who also received training) viewed all the structured interviews, which had been video recorded, independently and coded responses on response form I. The inter-rater agreement averaged 96% across all measurements (Bomman, 2001:5-3).These structured interviews were conducted in Eng lish.Although this was not the nurses' mother tongue, it was seen during the focus groups con ducted in the pre-experimental phase (Bomman & Alant, 2002:35) that their spoken use of English was good.It was assumed that this would not impact on the quality of the answers.All interviewers had access to response form I (con taining explicit guidelines) as well as their inter viewer notes to ensure that the procedure remained consistent.Eight open-ended questions were used for this purpose, e.g."I f you were the nurse work Table 1 Descriptive information on the participants (n=20)

Description Results
The nurses' ages ranged from 28 to 54 years and they tended to be older, with the majority being between 41 and 45 years of age.This highlights the fact that this is a group of adult learners and that adult teaching and learning styles had to be taken into consideration with the BCIP training.Only five nurses had less than three years experience (in cluding one with less than a year's experience) while three of them had more than 20 years experience.

Response fo rm II
This questionnaire had a total of 17 questions and five categories (biographic information, prior knowledge about disability and communication, attitudes and values, other positive outcomes and self-evaluation).Table 3 provides a summary of the most important areas covered.
Where possible, structured, close-ended questions were used.This format was selected to accommodate time con straints.Providing participants with a choice of possibili ties from which they had to select one, acted as a prompt and enabled the researcher to obtain an idea of what they knew, not merely of what they were able to write and spell.
During the pre-experimental phase it was seen that nurses feared to make spelling and grammatical errors.This type of format is also more motivational as it is easier and quicker to complete (McMillan & Schumacher, 2001:261).In addi tion, answers were easily scored and analysed according to pre-assigned codes.
The disadvantages of close-ended questions are that they are more difficult to construct and the fact that relevant answers can easily be overlooked (McMillan & Schumacher, 2001:261).In this research care was taken when reviewing the literature to determine possible answers; a pre-experimental phase was conducted (involving a focus group to highlight aspects that might not have been well described in the literature) and finally the questions were tested by means of a pilot study.
Some questions, e.g.Question 15 " When you think o f your skills as a nurse, which three things do you do best?"was asked in open-ended format, so that the participants were not guided in a direction, and also to allow them to answer in their own words.To a small extent, these questions were also used in an attempt to countercheck some of the other questions and to control bias in the development of the questionnaire (Brink, 1999:155).A few general guidelines were taken into consideration with the development of questionnaires (Brink, 1999:156;McMillan & Schumacher, 2001:258): • Items were mostly short and care was taken to en sure that each question dealt with only one con cept.

•
Care was taken to avoid any biased or leading ques tions, jargon and other difficult terminology, and to ensure the clarity of the questions, so that all par ticipants understood the same concepts with the vocabulary used.

•
An attempt was made to keep the questionnaires as short as possible to ensure that participants did not lose interest and/or become fatigued resulting in the omission of any important information.
• Meticulous care was taken with the visual appear ance of the questionnaires.Questions used during the different phases were printed on different col oured paper to make it visually more appealing.Questions were all numbered, organised in logical sequence and did not have too many items per page.
Care was taken to ensure correct spelling and gram mar throughout.• Clear, brief instructions were included at the top of the questionnaires to guide the participants.They were encouraged to complete all questions as in complete responses would impact on the reliability of the data.

BCIP training procedure
Training was conducted over five consecutive days, namely as it has repeatedly been shown that adult learners are more satisfied with learning if it applies to their everyday life and if it is practical and current (Givens-King, Sebas tian, Stanhope, Hickman, 1997:32).In addition, working in a problem-based manner with various case studies also ena bled nurses to extend and improve their knowledge base (in particular regarding disability and beginning communi cation skills), to remain contemporary in their field and to provide appropriate care for the unique problems they might face in their work (Jacobs, 1997:134).As it was expected of nurses to work independently after the training and to ap ply the principles to the various CSDs and their caregivers in their caseloads, the value of this approach is evident.
Other adult learning principles such as interactive work shops with activities that encourage a variety of forms of expression were also used.Activities such as role-play and observation elicited rich personal stories and concerns that, through facilitation, led to the development of a clear set of principles in completing case studies (Krogh & Lindsay, 1999:231).This technique was also helpful in de veloping problem-solving skills required to address antici pated or unexpected dilemmas.Although workshops, in order to reduce fatigue, never exceeded 60 minutes they were long enough to ensure that the nurses could master and practise the skills taught during the particular session.
After completion of the training all nurses received a cer tificate of attendance, leading to a feeling of achievement and general pride.An outline of the training is provided in Appendix I.

Results
Level 1: Cross-training of basic patient care skills This level refers to the medical handling of individuals through completing specific nursing tasks.Since it was not the focus of the training, this aspect will not be described further.
Level 2: Cross-training of professional non-clinical skills The nursing roles related to disability at this level include awareness raising, providing information, referral, feed back and follow-up as discussed in the introduction.Due Please note that each of the 20 participants were requested to provide 3 answers, hence the total of 60 items.
to the inter-relatedness of these tasks, they are described in a joint manner.
Results for this discussion were obtained from response form II, section E. The three open-ended questions dealt with a general self-evaluation of their skills as nurses; with those skills that positively impacted on the treatment of CSDs and their caregivers, and finally, with skills that they would like to improve in order to enhance their service de livery to CSDs.Combined frequencies for each of these questions were recorded for the different periods, and an item analysis was done.Table 4 depicts the nurses' selfevaluation of their general nursing skills during the three phases.
Table 4 shows that nurses became more aware of the impor tance of showing caregivers how to communicate with their CSDs (item 10).This awareness was maintained during the post-training and post-withdrawal phases.In addition, in the post-training phase they became aware of the impor tance of school placement and the provision of follow-up services.The greatest decline in scores during the study was noted regarding the counselling of caregivers to ac cept their CSDs and to adopt positive attitudes.This de cline may be due to the fact that, post-training, nurses had a concrete way of encouraging such acceptance (e.g.us ing the BCIP as opposed to the traditional counselling meth ods of talking and discussing).Some aspects received high scores throughout, such as general communication and listening skills, providing correct medical treatment and education through health talks.Referrals remained fairly Please note that each of the 20 participants were requested to provide 3 answers, hence the total of 60 items.
consistent over the three phases.
Secondly, nurses were asked to reflect on their skills that were particularly good when working with CSDs and their caregivers.The results are shown in Table 5.
Table 5 shows that, post-training, nurses were aware of the specific skills that were required when working with CSDs and their caregivers.For example, the highest score was obtained for "teaching caregivers to communicate with CSDs", and this high score was maintained during the post withdrawal phase.The other skills that were mentioned were specific in nature, for example: "using real objects to com municate", "demonstrating communication means and functions" and "providing communication opportunities".
As noted in Table 4, skills mentioned pre-training were vague and general, such as "teaching CSDs" and "facilitate a positive attitude by giving advice".These frequencies declined during post-training as attention became focused on specific aspects, such as "providing communication opportunities", "using aided communication" and "dem onstrating communication functions".During the post withdrawal phase, however, the frequencies for the general aspects increased again, but not to the pre-training level.
Results pertaining to the question about skills that the nurses would have liked to improve and/or receive training in, are shown in Table 6 .Please note that some scores are higher than 20.This is due to the fact that some aspects were grouped together, e.g.No. 7 included manual signs, EasyTalk, objects, etc. also interesting to note that "categorising CSDs accord ing to disability types" was no longer an issue, post-train ing.This is possibly due to the fact that the social model for disability was used (in accordance with the World Health Organization's classification system), where the emphasis is placed on participation rather than disability types (WHO, 2001:2).Post-training, nurses identified a whole new set of needs, with "teaching families to communicate with CSDs by using demonstrations" as the most pressing.This could possibly be attributed to the emphasis placed on social inclusion during the training.The other training needs re volved around the specific issues addressed during train ing; for example, expanding communication means and func tions and creating communication temptations.During the post-withdrawal phase the highest reported frequency was for "using different communication means".This might possibly be because nurses were already successfully us ing some manual signs (gestures for words like "more", "want", "give", etc.), and wanted to expand their current vocabularies.
Level 3: Cross-training of administrative skills At this level, prevention is the nursing role that relates to disability.The primary focus of the BCIP training was not the primary prevention of disability, but rather on second ary prevention, which is aimed at early identification and thus early referral for further assessment and treatment (Clark, 1996:450).This implies that the primary health care nurse adopts the role of early detector and referral source, by directing caregivers to other possible resources.Exam ples of such resources are the social worker (for informa tion regarding disability grants), genetic counselling (for family planning), and occupational therapy (for seating and mobility).In addition, secondary prevention highlights the primary health care nurses role as information sources and role models.The nurses model appropriate behaviours to families, while providing them with the relevant informa tion to enable them to make appropriate and knowledge

Summary of results
Means that do not differ statistically significantly from each other, are underlined.
able decisions about their children's rehabilitation (Roberts, Rule & Innocenti, 1998:69).The type of advice provided to caregivers following a specific case study was obtained from response form 1, section C. Results are shown in table 7.
Table 7 shows that nurses became more aware of different kinds of advice that could be given, as seen in their total scores.Pre-training advice tended to be general (e.g."coun sel caregivers on acceptance").Post-training and post withdrawal, the advice given tended to become more spe cific in nature (e.g."increasing communication means" and "increasing social interaction").Generally, nurses also tended to become more aware of the importance of referral to a mainstream school (i.e.referral to special schools de creased).The decline in "referraF' over the various research phases was also interesting.This could possibly be due to the fact that nurses became more empowered and confi dent in assisting CSDs and their caregivers.
Level 4: Cross-training of clinical disciplines At this level, the nursing roles that relate to disability con cerns early identification and screening and planning serv ices.Simply locating and identifying CSDs is not enough; services aimed at minimising the effects of a disability should also be investigated.The purpose of identification is thus intervention (Wilkey & Gardner, 1999:303).The first aspect of identification and service planning entailed making nurses more aware of the available services and how they could assist caregivers to access these services, and of the importance and role of complete case histories and the pro vision of specific information.Results related to this as pect are discussed in tables 4 to 7.
Once the disability has been identified, service provision should commence.CSDs and their families often live in remote rural areas, resulting in primary health care nurses being the only professionals available who are able to pro vide co m p reh ensive, h o listic services "on the s/?oi" (Lequerica, 1997:287).In order to achieve this ideal, professional training should be expanded so that primary health care nurses are able to provide services, albeit lim ited, to the caregivers of CSDs.This is the nursing role for which primary health care nurses required the most input, and consequently, the emphasis of the BCIP training fell on the provision of knowledge, which was intertwined with skills development.1985:3).
Results pertaining to applied knowledge were recorded on response form I, section B. A Friedman test was employed to determine whether the change in applied knowledge was statistically significant over time.A r-value of 0.0001 was noted, implying a statistically significant difference.A sum mary of these results is shown in Table 8.The pair wise comparisons showed a statistically significant increase in global applied knowledge at the 1% confidence level, be tween the pre-training score (O,) and all the following scores, namely the post-training scores (0 2), the two-week follow-up scores (0 3), the six-week follow-up scores (0 4), and the post-withdrawal scores ( 0 5).No differences were noted between O,, O,, O, and O,.To yield data regarding their skills, nurses were asked to demonstrate service delivery in a number of different com parable cases.These demonstrations were video-recorded and documented on response form I, section D. Two inde pendent raters rated the results according to a number of different variables.A 96% inter-rater reliability was noted throughout (Bomman, 2001:5-3).A Friedman test was em ployed to determine whether the change in skills was sta tistically significant over time.A r-value o f0.0001 was noted, implying a statistical significant difference.A summary of these results is shown in Table 8 .It indicates a statistical significant increase of combined skills at the 1 % confidence level between the pre-training scores (Oj) and all the fol lowing scores, namely the post-training scores (0 2), the two-week follow-up scores (0 3), and the post-withdrawal scores (Os).Results also indicated a statistically signifi cant increase at the 1 % confidence level between the post-training (O.,) and the six-week follow-up scores (0 4), and between the post-training (0 2) and the post-withdrawal scores ( 0 5).This implies that skills increased significantly in statistical terms during the follow-up periods, compared to pre-training and post-training.This emphasises the im portance of follow-ups in skills development.

Conclusion and recommendations for further research
This article aimed to provide a theoretical framework for understanding the pivotal role of primary health care nurses in providing services to CSDs and their caregivers.Multi skilling, an in-service training strategy through which pro fessionals expand their knowledge and skills, was explored.Finally, research results following a particular multi-skilling training programme, viz.BCIP training, were analysed ac cording to the various nursing roles of primary health care nurses while working with CSDs and their caregivers.From this analysis it became clear that the training had a signifi cant impact on primary health care nurses' acquisition of knowledge and skills in the field of severe disability.
Although this article discusses the value of multi-skilling, it also points out certain gaps in the existing body of knowl edge.Recommendations for further research are provided.
As discussed, a comprehensive approach to PHC aims at equipping personnel (primary health care nurses, in this research) with many different skills.This multi-skilling could lead to nurses feeling insecure about their primary roles and responsibilities, which could greatly impact on their job satisfaction and motivation.The relationships between multi-skilling and job satisfaction, between multi-skilling and overburdening, and between multi-skilling and moti vation to work with CSDs and their caregivers should be further investigated.A fter the discussion and demonstration of a particular communica tion function and temp tation, participants were asked to practise the skill using the BCIP.Fi nally nurses were di vided into small groups and each group had to explain how they would conduct service deliv ery and then dem on strate their newly ac quired skill.All case studies differed to en sure that there was no duplication of facts and that all the different fu n ctio n s th at w ere taught could be demon strated.An example of a case study is: "Simon is a fo u r year old spastic boy.The four main communica tion dom ains were d is cussed, h ig h lig h tin g means.This was followed by a discussion of the con cept "Augmentative and Alternative Communica tion'" after which the aided and unaided strategies in cluded in the BCIP were d iscu ssed and dem on strated.The unaided strat egies included gestures (e.g.p o in tin g , yes/no headshake, facial expres sions, miming and ges tures) sign language (SASL) and natural ges tures.The aided strategies include real objects, colour photographs, PCS and a 4option digital speaker.Ad vantages and disadvan tages of each were high lighted.
After a discus sion and demonstration of the various aided and unaided communication strategies included in the BCIP, nurses had the opportunity to prac tise these skills.All means were first prac tised in isolation (e.g."Make the gesture fo r "o p e n ") after which they had to incorporate it with their skills of the previous day (e.g.use a gesture to provide a choice).This was done is small groups where the roles co n stan tly changed betw een a nurse acting as the CSD, a caregiver and a nurse.Skills were then applied to a specific case.4 1.To revisit the four major areas that impact on com munication 2. To discuss different com munication contexts / envi ronments 3. Three video cases were shown and a Progress C h eck list was co m pleted for each case.This was done in the group as a whole.Each score was then d is cussed in order to gain consensus and to act as in-service training.This was followed by another two more video cases, w hich nurses completed a checklist in the sm all groups.Scores were again dis cussed and nurses were able to objectify their given scores.High cor relation was found after the 5th video case.This was followed by a case study discussion simi lar to the one discussed in response form I.

Figure 1 :
Figure 1 : Roles of primary health care nurses within multi-skilling levels

Figure 2
Figure 2 Six experimental phases consultation 2 weeks post-training: Assessing skills of com munity nurses.Identification of problem areas and problem solving.Data obtained by means of response form I. Phase 0 4 Follow-up consultation 6 weeks post-training: Assessing skills of com munity nurses.Identification of problem areas and problem solving Data obtained by means of response form I. Phase O s Follow-up consultation 5 months post-training: Assessing skills, attitudes and knowledge of community nurses.Data obtained by means of response form I and II.skilling dimensions relevant to primary health care nurses when working with CSDs and to discuss the outcomes fol lowing training aimed at multi-skilling Level 2 (cross-train ing of professional non-clinical skills), Level 3 (cross-train ing of administrative skills) and Level 4 (cross-training of clinical disciplines).
6 -10 11-15 16-20 >20 Years of experience each of the 20 participants were requested to provide 3 answers, hence the total of 60 items.M onday to Friday at the U niversity o f P re to ria 's Hammanskraal campus in the Moretele district, as it re stricted travelling time and costs for the nurses.Principles of adult learning were used throughout the training.Each day the training started with a brief revision of the previous day's work.The nature of the training was problem-based

Table 6
yields some interesting results.Pre-training, eight een, thirteen and ten nurses respectively required more train ing in "training caregivers and siblings", "interviewing caregivers and communicating with them" and "basic training o f CSDs".It is clear that the BCIP training pro gramme addressed these issues, because a noticeable post training decline was observed for these three aspects.It is

Table 8 Friedman test of knowledge and skills
As expressed so aptly by Bruner, "It matters not what we have learned.What we can do with what we have learned; this is the issue... " (cited in Brewer,