• surveillance;
  • resistance;
  • health needs assessment;
  • health visiting;
  • medicalisation;
  • conversation analysis;
  • Foucault


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Abstract  This paper draws on Foucault to understand the changing discourse and impact of structured ‘health needs assessments’ on health visiting practice. Literature about this activity makes little mention of the long-standing social purposes of health visiting, which include surveillance of vulnerable and invisible populations, providing them, where needed, with help and support to access protective and supportive services. Instead, the discourse has been concerned primarily with an epidemiological focus and public health, which is associated with risk factors and assessments.

The use of pre-defined needs assessment schedules suggests that health visiting activity can be sanctioned and clients’ needs serviced only if they reach the threshold of pre-determined, epidemiologically-defined risk. Their effect on practice is examined through a conversation analysis of ten health visitor/client interactions using two different structured needs assessment tools. The study indicates that the health visitors, like their clients, were controlled by institutional expectations of their role; analysis of their conversations shows how they achieved the requirements of the organisational agenda. Structuring client needs and health visiting practice through the use of formal needs assessment tools emphasises the epidemiological focus of the health service above the need to arrange support for vulnerable individuals. In this respect, it serves as a marker in the continued medicalisation of health visiting.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

This paper describes a study about the use of structured health visitor needs assessment tools in two different parts of England. A conversation analysis of health visitor/client interactions, in which structured instruments were used, provides a base from which to examine the changing discourse and impact of ‘health needs assessment’ on the health visiting service.

Health needs assessment is implemented in various ways, but is commonly used to justify the restriction of services to some clients and delivery to others. In this respect, it illustrates a mechanism through which society legitimises public funding for some forms of support whilst others are not approved. Health visiting has a long history of being a women's service for women (Dingwall 1977, Dingwall et al. 1988). Traditionally, the work has been loosely structured, operating through systems of support, friendship and inter-personal relationships to improve the health of, predominantly, infants, young children and their mothers (Dingwall 1977, Davies 1988). The focus on social surveillance and child protection was an implicit theme throughout the 20th century (Abbott and Sapsford 1990, Dingwall and Robinson 1990, Peckover 2002). Health-enhancing behaviour changes were sought through the informal transfer of knowledge and information, with practitioners maintaining a broad agreement for their purpose by operating within a purposefully ambiguous and shifting conversational context (Dingwall 1982, Cowley 1991).

The writings of social philosopher Michel Foucault (1976, 1979, 1984) help to explain the apparent compliance of clients and acquiescence of health visitors in the implementation of the controlling and potentially disempowering assessment processes identified in the study reported in this paper. The inhibition of client participation by health visitors has been reported before (Sefi 1985, Kendall 1993), as have concerns about the restrictive impact of using population-wide, normative needs assessments as a means for targeting health visiting intervention (Dingwall and Robinson 1990). It is argued here that both these negative effects are brought together and exacerbated by the implementation of structured health needs assessment tools into everyday health visiting practice, which serve as a significant marker in the continued medicalisation of health visiting.

Background to the study

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

The central premise of Foucault's work lies in the relationship between knowledge and power, and the way these are used in the language of individuals and groups to construct a social reality (Foucault 1976). Interaction is fundamental to this central premise, which sees knowledge and power relations as components of discourses. ‘Discourse’ refers to the phenomenon whereby, through the use of language, individuals construct a social reality which, in turn, reinforces power relations. This construction of social reality reflects and reinforces the power relations that exist, in this case between health care professionals and their clients. Thus, Foucault characterises power as part of the process of interaction.

A number of scholars have analysed health visiting work in Foucauldian terms, often focusing on the dichotomy between the policing and supportive elements of the role (Abbott and Sapsford 1990, Bloor and McIntosh 1990, Dingwall and Robinson 1990, Heritage and Lindström 1998). Some of these authors have concentrated on the controlling aspects of the role, reporting the resistance tactics of the women being visited (Bloor and McIntosh 1990, Peckover 2002). Others have emphasised the perceived social importance and wide acceptance of surveillance aspects of the role as a necessary means of ensuring that vulnerable children and others are protected (Dingwall 1982, Dingwall and Robinson 1990).

A small number of studies using conversation analysis (CA) to examine interactions between health visitors and their clients, show both similarities and differences with the pattern of doctor/patient interactions revealed in a review of the CA research by ten Have (1991). Describing the asymmetry in such interactions, ten Have describes the ‘natural’ dominance by the physicians, as they gradually establish topic relevance, asking a series of short, unexplained questions to establish the biomedical ‘facts’ of the case, whilst selectively ignoring aspects that, mainly, relate to subjective patient experience. The patient usually acquiesces in the doctor taking the lead in questioning, using a clear information-gathering format. Although there is asymmetry, the conversational tasks of both interactants contribute to achieving the (usually) mutually-desired endpoint of the ‘medical decision’.

In most doctor/patient consultations, the patient initiates the encounter. This stands in contrast to health visiting visits to the home, which are generally unsolicited. It is usual, now, for contacts to be prearranged, possibly as the result of a mother's request, but two significant CA studies (Dingwall and Robinson 1990, Heritage and Sefi 1992, Heritage 2002) collected data when it was common for health visitors to turn up unannounced; both studies focused on a ‘new birth visit’. The National Health Service (NHS) is required to provide a free maternity and child health service and this visit usually marks the transfer of responsibility from the midwife, who has primary responsibility for the woman in pregnancy, during and after delivery, to the health visitor, who traditionally maintains contact with the family until the child starts school. As in the doctor/patient conversations, asymmetry between health visitor and client occurred in both studies, with Heritage (2002) reporting the completion of ‘face sheet’ data on the birth record as a series of short questions and answers that are similar in character to those found in ten Have's (1991) review of medical consultations. Whereas the patient-initiated start of the doctor/client encounter was straightforward, Dingwall and Robinson (1990) focused on the health visitor's entry to the house as a key point for analysis. This illustrated the health visitor's expectation of access and co-operation, framing the rest of the visit and interaction. Beyond that, Dingwall and Robinson (1990) noted that the interaction is like an ordinary conversation which, unlike the medical consultations, followed no discernible or overt agenda. The content was not designed in advance but ‘rather, at each decision point (at every turn), continuation was organised to the parties’ satisfaction by a selection from the range of options seeming to be legitimately available’ (1990: 263). Any specific purpose seemed opaque to the researchers and, possibly, to the mothers receiving the visits, although the surveillance aspects of the home visit, as explained above, were achieved.

In the last decade, there has been a dearth of research or papers commenting on the child protection functions of health visitors. This silence on the subject runs counter to the experience of most practising health visitors whose daily work is still taken up with much child welfare work. However, it reflects the official rhetoric, which is that child protection duties are primarily the concern of social, not health, services. Referring to the traditional difficulty of reconciling their child protection responsibilities with community-wide activities, for example, the Department of Health (2001) urges health visitors to ‘play a more proactive role in promoting the welfare of children using a public health approach . . . [ . . . ] . . . In the context of child protection, this means looking at risk patterns in a community to identify children and families potentially in need of support . . .’ (2001: 30). The official version, therefore, recognises the importance of child abuse as an epidemiological risk factor that has implications for public health. Public health, not child protection, is considered the major function and justification for health visiting, now, and the former emphasis on relationship-building has been largely displaced in official versions of the role. In line with the epidemiological focus of public health, the discourse has shifted within health visiting, as in the NHS as a whole, towards the process of a general ‘health needs assessment’ as a basis for targeting preventive services most effectively.

Health needs assessment tools (HNAT)

Given their focus on identifying and supporting vulnerable populations, the ‘search for health needs’ is acknowledged as a basic principle of health visiting (Council for the Education and Training of Health Visitors 1977, Chalmers 1993). The means by which this is achieved, or the reason for it, has been unclear to both observers and users of the service. Concern about the seemingly unfocused and uncontrolled nature of health visiting services led to a degree of censure in the early 1990s (Audit Commission 1994, Roberts 1996). In a policy guidance, the NHS Executive (1996) emphasised that health visitors, like others, should target their practice according to identified need. In response to this expectation, a plethora of clinical practice guidelines and assessment tools emerged to assist practitioners in the identification of vulnerable families that require targeted health visiting interventions.

During 2001/2002, the NHS Community Trusts that employed health visitors were replaced by Primary Care Trusts, that commission many hospital-based health services as well as directly providing community services such as district nursing and health visiting. A national postal survey of the former Community Trusts in England revealed that 63 per cent of them issued health visitors with some form of official practice guideline to assist them to identify those families in greatest need of the service (Appleton 1997). The majority of these guidelines were classified as checklists, scoring systems and screening tools. Despite the formal position that health visiting is no longer concerned with child protection, most of the guidelines appeared to be heavily influenced by the risk assessment approaches used in that field.

Appleton and others questioned the value of such guidelines, suggesting that they were often invalid, which may lead to families being wrongly classified (Barker 1990, 1996, Browne 1995, Goddard et al. 1999). Critics of structured approaches to needs assessment claim that this approach interferes with development of a necessary health visitor relationship (Houston and Cowley 2002). Needs assessment should, it is suggested, be seen as a complex, skilful activity, requiring professional judgement (Appleton and Cowley 2003) and the ability to adopt an empowering approach that enables client participation (Machen 1996, Normandale 2001, Bidmead et al. 2002).

Set against these criticisms, advocates of structured assessments emphasise that they are intended to be used in partnership with the client (Browne 1995, Crompton et al. 1998), who can be ‘talked through’ a list of questions, prompts and triggers that will help them think about needs that they may not otherwise have considered. The professional judgement of the practitioner, it is claimed, can still be afforded a greater part than scores identified by use of the tool (Crompton et al. 1998), whilst improving the transparency of the service may enhance client control over the process. Furthermore, of importance to managers and commissioners, such proforma provide a mechanism for keeping the service within bounds, rendering it accountable by listing specific, medically-recognised needs that merit attention.

This emphasis on the mechanics of needs assessment masks wider concerns about the impact such requirements may have on the health visiting role or practice. Whilst a few papers describe developing the structured approaches to needs assessment (Browne 1995, Crompton et al. 1998, Naughton and Heath 2001), there is far less research about how their use actually works out in practice. The extent to which a structured interaction enables clients to participate in defining their own needs assessment process was largely unknown, as was the effect of the instruments on health visiting practice.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

The aim of this research was to examine what happens between the health visitor and client during an assessment process using a structured needs assessment tool. The responsibility of the health visitor to assess the needs of the family is discharged through verbal, face-to-face interaction with clients. Foucault (1979) characterises power as part of the process of interaction, being ‘exercised at innumerable points in the interplay of non-egalitarian and mobile relations’ (1979: 164). A research approach was needed that would capture these interpersonal relations between the health visitor and the client during the assessment process, whilst acknowledging the wider organisational influences upon the conversation. There are inherent tensions in such an endeavour, although Miller and Silverman (1995) maintain that their blending of conversation analysis and a Foucauldian perspective made it possible ‘to analyze how power relations are both embedded in institutional discourses and constructed within social interactions’ (1995: 726).

This study followed the example of earlier research on health visiting interactions (Sefi 1985, Robinson 1986), in using the conversation analysis (CA) approach developed by Sacks, Schegloff and Jefferson (1974) and defined as ‘the study of talk-in-interaction’. This is part of a broad tradition that stresses the functional, action-orientation of discourse, by looking in detail at the organisation of social interaction and what it is designed to accomplish (Gill 2000). The method is based on the assumption that it is fundamentally through interaction that institutional imperatives originating from outside the conversation are produced and made real and enforceable for the participants.

Ethics committee approval was obtained, and data gathered from volunteer health visitors in two NHS Community Trusts. Site A is in a large city, whilst Site B covers a mixed urban/rural area. Each health visitor was asked to select a client with whom she/he planned to carry out a home visit for the purpose of needs assessment and to tape record the interaction. Given the intensity of the analysis, ten health visitor/client interactions, five from each site, were deemed an adequate sample. Assurances of confidentiality were given to both health visitors and clients and informed, written consent to participate was obtained.

The health visitors in Site A were required by the employing trust to use a highly structured assessment questionnaire and complete it within 16 weeks of the initial visit, which was usually soon after a new baby was born. However, given the size and nature of health visitor caseloads in this Trust, in practice the assessment was usually accomplished within one or two visits. The assessment resulted in families being scored as having high, medium or low needs, which was entered on a database. Although health visitors were involved in the working party that developed the assessment tool, there had been considerable controversy in the Trust when the health needs assessment tool (HNAT) was introduced, leading to serious dissent between managers and some of their staff. Accordingly, the process was formally audited, and health visitors believed that failure to comply could result in their being subject to disciplinary action by their employers. Officially, clients were allowed to opt out if they wished.

In Site B, health visitors were required to use the early intervention strategy (EIS), a framework developed elsewhere by Crompton et al. (1998) and adopted by the Trust, to guide their assessment. The EIS was originally developed over a period of five years and has its roots in child protection (Naughton and Heath 2001). In practice the health visitors were to use the EIS tool to screen all families within the antenatal period and, if concerns arose, a second in-depth assessment followed. Within this Trust, some health visitors expressed reservations at the start, but the requirement to use the EIS guideline was adopted without the level of controversy reported in Site A. There seemed to be limited auditing of practice and, at the time of the study, no real compulsion to comply.

The data were collected by means of direct audio-taping of natural practice, controlled by the health visitors who could choose which interactions to record, in anticipation that they would record what they believed accurately represented health visiting practice using the structured assessment tools. In Site A, the tape-recordings were part of a wider study (Cowley and Houston 2003) in which the five health visitors were selected from 30 volunteers, to represent a range of views about their health needs assessment tool. Site B was identified specifically to complement these data, being a Trust which, despite reservations on the part of some health visitors, had implemented a different, but similarly structured, HNAT without any overt controversy among the staff. Volunteers were sought from two separate parts of the Trust, that had each introduced the EIS in a slightly different way and time period.

The tape-recorded interactions of the administration of the HNA tool were transcribed using the Jeffersonian Transcription System, which was adapted to the purpose of the study and its potential audience (Psathas and Anderson 1990, ten Have 1999). To that end, the transcriptions include nuances, pauses, laughter and non-committal utterances, for example, ‘umm’ and ‘err’, but are not detailed transcription notations, as the primary purpose of the research was to understand health visiting practice rather than the organising features of conversations. Nevertheless, it seemed important to utilise the transcription symbols and conventions characteristic of conversation analysis, so that potential CA readers of the study would perhaps find it more meaningful.

The institutional nature of the interaction was probed by analysing the data, using the six key points detailed by Heritage (1997). These focus, first, on the existence or not of a specific turn-taking system within the interaction, then establishing a ‘map’ of the interaction in terms of its phases and how these are developed; this reveals the overall structural organisation. This kind of sequence organisation lies at the heart of conversation analysis, as it shows how opportunities for action are opened up and activated, or withheld and occluded. ‘Turn design’ within the conversation refers to two distinct selections that a person's speech embodies: the action that the talk is designed to perform and the means that are selected to perform that action. The choice of words used, such as the organisational we or institutional euphemisms are components of turn design . These ‘lexical choices’ are significant ways of talking that may be used to minimise discomfort provoked by the situation or topic under discussion. Finally, interactional asymmetries are noted; they may be manifest in a number of ways, particularly in the asymmetry of participation, knowledge and know-how in the conversation.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

In both sites, there were signs that the health visitors departed from the officially prescribed method of using the assessment tools. The departure was most marked in Site A where their managers attempted the firmest control over the health visitors. Here, the health visitors went against instructions to avoid using the tool as a ‘checklist’ of questions and administered it as a survey, rendering the assessment and the organisational agenda very overt. In contrast, in Site B, instead of using the instrument as a tool to guide partnership, it was generally implemented covertly, so the organisational agenda was kept firmly hidden. These differences may have been attributable to the differences in the instructions they were given, or in the form and structure of the different needs assessment instruments.

Despite, however, some differences in the way they resisted the prescriptions on their practice, health visitors in both sites showed some marked similarities in the way they implemented the actual needs assessment process, so it seems most likely to be attributable to the structured nature of the instruments they were using. In both sites, the assessment process was structured so that the health visitor controlled the interactions, and there was limited opportunity for the client to participate in identifying or determining their needs. The findings are presented in three stages that reflected the way the HNATs were implemented, first by introducing the assessment tool, then the process of the assessment and its outcome in determining the level of need and planning future interventions. The process of the assessment involved both information gathering and information giving, which are considered in more detail elsewhere (Mitcheson and Cowley 2003).

Introducing the health needs assessment tool

The analysis revealed significant differences between sites A and B in the initial phase of the interaction, when the health visitors introduced the assessment format to the client. Health visitors from Site A structured the interaction so that there was a clear demarcation between the assessment process and the rest of the visit where, on the whole, the health visitor focused on what might be considered routine aspects of health visiting practice about maternal and child health, as shown in Extracts 1 and 2:

Extract 1 HV/C2A

Line 1 HV When I finish this we can deal with you know other things when

Line 2 we've finished with the questionnaire we can deal with (.) you know

Line 3 aspects (.) any questions you want to ask about the baby.

Line 4 Client Yes, that's fine.

Extract 2 HV/C3A

Line 146 HV right I've done that bit there

Line 147 So:: lets go through your little heath visiting assessment tool then

Generally, the health visitors in site A explained the assessment process to the client in depth (Extract 3):

Extract 3 HV/C2A

Line 11 HV And basically it's a format that ermm . . . we started using in last year (.)

Line 12 ((trust name)) May 1999 and it's rather than us

Line 13 just coming in, you know, asking standard questions about how

Line 14 you: are etcetera=

Line 15 this is a special format that sets up (.) you know sort of

Line 16 headings it sort of you know jogs your memory of anything

Line 17 you want to discuss yeah?

Line 18 Client OK

On occasions, the health visitor launched into the questions without any prior introduction or information regarding the purpose or nature of the assessment. This approach may stem from an assumed right of the health visitor to use her institutional position to ask a series of assessment questions; the client appears to concur by responding. Even so, the HNAT used in Site A was very highly structured and health visitors generally dissociated themselves from the assessment format, often through the use of institutional euphemisms such as we as indicated in line 11 of Extract 3, above. This provides a clear illustration of the way, noted by Sacks (1992), in which potential difficulties for the institutional representative can be minimised. Hesitancy, talking in whispers, frequent pauses and the use of utterances such as ‘ermmm’ all appeared throughout the HNA phase from Site A. These conversational indications that the health visitors felt uncomfortable with the line of questioning they were about to embark upon were absent except when the HNA was being used.

In site B, the demarcation was much less noticeable and the phases of assessing need could only be clearly identified in one interaction (Extract 4), when the health visitor informed the client about the EIS at the start of the visit; once more, the hesitancy is apparent.

Extract 4 HV/C4B

Line 3  HV okay ermm

Line 4 we (.) we visit all first time mums for an ante-natal visit ermm but

Line 5 over the last year things have been changed slightly=

Line 6  Client right

Line 7  HV in that we’re now doing what's called the Early Intervention

Line 8 Strategy which is sort of a bit of a longer visit really so//

Line 9  Client //ummm

Line 10 HV to sort of get to know you:: and your family before your baby

Line 11 arrives

Line 12 Client umm umm

Even though the health visitor introduces the assessment to the client in Extract 4, the specific reason for the assessment, which is to identify risk factors for child protection, is obscured by the description at lines 8–10, where the health visitor refers to a longer visit to get to know the client. It would appear from this observation that some health visitors are assessing the needs of their clients covertly, which is likely to lead to some communication difficulties. This reflects the general obscuring, at a national level, of health visiting responsibilities towards child welfare. Nevertheless, in institutional interactions, where the objectives of the encounter are unclear, opaque or even suspicious to one or both of the participants, then confusion and conflict of interests are more likely to occur.

The process of health needs assessment

The main body of the needs assessment process is accomplished primarily by information-gathering, through long question-and-answer sequences of interaction. This kind of sequence creates a strong moral expectation that each question will be followed by an answer and when it does, it gives control back to the questioner (Perakyla and Silverman 1991). Significant asymmetries of participation occurred in both sites, with more questions overall being asked by both interactants in Site B. In Site A, the questions are unequivocal and short, designed specifically to gather information, with yes/no answers being prominent. This kind of asymmetry has been widely reported in doctor/patient interactions (ten Have 1991) and health visitor/client ones (Sefi 1985, Kendall 1993).

Where a client was able to raise her concerns, as in the following extract, it was invariably blocked or minimised by the health visitor:

Extract 5 HV/C5A

Line 203 HV How happy are you with ((child's name)) behaviour? 

Line 204 (.) Very happy, fairly happy, not happy?

Line 205 Client I guess fairly happy ermm I would say very happy

Line 206 HV Does she wake a lot?

Line 207 Client I think she is pretty good in that sense.

Line 208 she was a little bit (.) I would guess colicky in the first two weeks

Line 209 but that has settled down now and she's now trying to show that she can

Line 210 sleep through the night.

Line 211 She had a ten hour period, two five hours back to back with a feed

Line 212 between which lasts about (( )) and then she wakes about 5 or 6=

Line 213 We’re just having a hard time getting her down for that initial sleep

Line 214 because she likes to be on you (.) she doesn't like to be put down

Line 215 HV We can sort of discuss that at another time

Line 216 And what about yourself, how are you in yourself?

Despite the client's attempts to inform the health visitor of her needs, it seems the health visitor will not or cannot be distracted from her line of questioning. This appears to be determined by the assessment format, which requires a decision between the choices offered at line 204. Although she informs the client that she will return to the problem of settling the baby she does not, at least not during this visit. Instead, at line 216, the health visitor signals that the topic is closed and through the use of ‘and’ that a new one, the next on the list of questions, is beginning.

This was a recurring pattern throughout the assessment phase of the interaction in both sites but was most marked in Site A; the identification of a source of anxiety or concern by the client would only be approved as a ‘need’ if it accorded with the prior formulation given in the assessment tool. Indeed, several health visitors continued the line of questioning until the client gave an answer that fitted the categories available to them on the assessment format. In Extract 6, for example, at line 40 the client agrees with a deep sigh, perhaps indicating that she feels a pressure to give a clear answer, instead of opening the issue for discussion or dismissing it with humour, which were possible options suggested at line 37:

Extract 6 HV/C1A

Line 36 HV Is this a good time for you to be a parent?

Line 37 Client When isn't, when is? ((laughter))

Line 38 HV yeah::

Line 39 So are you saying yes, or no?

Line 40 Client ((sighs)) ye::ah

In Site A, service managers envisaged that the tool would be used to guide assessment, rather than as a checklist of questions, which is how the health visitors used it. Similarly in Site B, lengthy sequences following a checklist type of format could be identified in all the interactions. Instead of simple yes/no responses, however, sequences tended to be more exploratory in nature, enabling clients to express their own perspective more readily, as in Extract 7:

Extract 7 HV/Client 2B

Line 264 HV And (.) is that how your childhood was too

Line 265 Client yeah:

Line 266 HV your Mum was really good with you

Line 267 Client yeah she used to take me out//

Line 268 HV //yeah

Line 269 Client to the park and that every day

Line 270 HV yeah so you tend to look at your Mum and think oh I'd like to be like my

Line 271 mum

Line 272 Client yeah yeah

Line 273 HV What about erm (partner's name) did he have a happy time as a child

Line 274 Client oh yeah yeah he had erm a brother and three sisters what (.) which are older

Line 275 than him you see so he sort of (.) he had (..) well what his mum told me

Line 276 he he had a happy sort of life and that you know

Line 277 HV It does make a difference you know because (.)

Line 278 Client oh yeah he's a brilliant dad you know

In both sites, it was common for the health visitor to inform the client of the reason for her question after the client had answered, as in Extract 8.

Extract 8 HV/5A

Line 89 HV Have you or (partner) experienced any mental health problems

Line 90 (.) psychological problems depression?

Line 91 Client No.

Line 92 HV The reason around that especially for you is to do with postnatal depression

Line 93 because in a way it can start from the time of birth up to eighteen months of

Line 94 age

Sometimes, instead of this kind of explanation about the direct purpose of the question, it appeared that the health visitors were making some kind of moral justification for requiring the information, as in Extract 9.

Extract 9 HV/C 4B

Line 275 HV we have to ask some quite deep questions

Line 276 and (..) the reason we ask those questions is not because

Line 277 we’re nosy not because we’re prying it's because (.) you know the

Line 278 more we know about strains within the family then the more we

Line 279 can help

Here, the option to withhold the information is effectively closed down, with the health visitor leaving the client in little doubt that she requires the information and therefore the client has an obligation to provide it. By responding positively, as shown in Extract 7, the client accepts the opportunity to display, in Dingwall and Robinson's (1990) terms, ‘their good moral character and receive public validation for it’ (1990: 258). Extracts 7 to 9 vividly illustrate Foucault's (1984) concept of ‘pastoral power’, which is one element of the wider concept of bio-power. By listening, caring and encouraging the exposure of intimate ‘confessions’ professionals, like the health visitors here, can widen the focus of their surveillance into the most intimate part of their clients’ lives.

Overall in Site A, the questions followed the exact format of the questionnaire, so the interactions showed a marked uniformity, in the manner of recording a medical history. Stylistic variation was more evident in site B where the sequences were more vague and difficult to identify. Although the topics covered were similar and mirrored those in Site A, the line of questioning was generally more open and allowed for the client to offer her/his perspective more readily. Nevertheless, the objectives appeared to be the same in both sites: to gather information about a predetermined range of issues and to begin the process of informing the client about which needs the health visiting service was there to meet.

Outcome of health needs assessment

The officially-stated purpose of the assessment is to determine the nature and level of need of the client, so the health visiting service can be targeted effectively. The outcome of the assessment, therefore, included two stages: determining the level of need, then closure and planning future interventions. The key difference between the two sites, the degree of transparency about the whole process of needs assessment and its purpose, became very clear at this point. In site A, the decisions were mainly explicit and overt, whilst in Site B, a covert process was more usual.

Examples of the ways in which the health visitor and client determined the level of client need varied slightly in the extent to which the client was enabled to participate. In extract 10, for example, the health visitor ensures that the client is informed of how her level of need is determined. She explains the task so that the client, too, follows the ‘approved’ decision-making process:

Extract 10 HV/C2A

Line 243 HV What we do now (.) at the end of all that (º)

Line 244 we look at the score you know whether how many yes's (º)

Line 245 erm and we have high, medium or low category yes?

Line 246 Client okay

Line 247 HV and its is not for us to decide which category you want to be

Line 248 in, but it is up to you to say which category you feel you come into

Line 249 from all the questions and from all the answers you've given me.

Line 250 Client okay=

It would be tempting to conclude that the client is participating in determining her level of need but it is debatable whether she has played an active part in the main decision-making process. Indeed, for some clients the idea that the aim of the encounter had been to classify needs seems to come as a surprise, as indicated in Extract 11 when the health visitor, having completed the HNA, is required to determine the level of need of the client as high, medium or low:

Extract 11 HV/C1A

Line 83 HV this is the bit of the time when I sort of . . . I mean how would you

Line 84 classify your needs?

Line 85 Client my needs?

Line 86 HV Your needs yes (.) er I suppose immediately the baby came home it was

Line 87 a bit high need there wasn't it? Yeah

Line 88 Client yeah I was asking people cos she's got chest problems (.) then when

Line 89 I went down the clinic they gave me a number to ring someone down at

Line 90 (local) Council to do with getting a grant, but cause there were no

Line 91 medical grounds so she put me through to someone else which I said it was

Line 92 for medical grounds and she ain't rung me back so (....)

Line 93 HV That would be great if you could get central heating but anyway . . . I mean

Line 94 I would put you as a medium at the moment because you’re still in that (.)

Line 95 I know you’re actually quite anxious about the baby and er (.)

Line 96 we need to get something sorted out with the heating.

Line 97 Client Yeah . . .

Here, the client appears to have only a dim awareness of the professional objectives being pursued across the entire encounter. A lack of knowledge may mean that clients struggle to understand the purposes lying behind particular questions and may not grasp the line of enquiry that the professional is pursuing on what might appear to be unconnected topics (Heritage 1997). It is not difficult to see how such interactional asymmetries of ‘know how’ have the potential to be a source of difficulty and confusion for the client and are unlikely to result in the client being able to identify her/his needs effectively. The health visitor responds to the expressed surprise by giving ‘approved examples’ of what might constitute a need in line 86 and 95, but does not explain why the needs assessment is taking place in the first place.

In most cases in Site B, no feedback was given to the client of the health visitor's assessment of the client's need. This is not surprising in light of earlier comments that, in the majority of interactions, there was no indication to the client that the purpose of the visit was to assess their need for the service. Where any mention was made, there seemed very little evidence of the client participating in determining their level of need other than to confirm the health visitor's assessment, as in Extract 12.

Extract 12 HV/1B

Line 372 HV I’ll leave you my card and phone number

Line 373 and then what I do is I've just ticked on here ermm core health visiting

Line 374 package. I mean I don't think there is anything that you particularly need

Line 375 is there? I mean it seems as if you've got a lot of help and everything's

Line 376 all ready

Line 377 Client yeah

Line 378 HV ermm so that's fine

Line 379 Client Okay.

In line 374, the casual wording of the remark that ‘I don't think there is anything that you particularly need’ rather obscures the fact that this is an official assessment, even though the decision to offer a minimal (core) service is recorded on the form. That decision is conveyed as a statement (line 373), rather than as an option to be discussed. Slight though it is, this is the only example in Site B of any explanation about the assessment process. The client indicates acceptance of the health visitor's decision, confirming her position as a passive recipient within the process.

In theory, clients could choose whether or not their needs were assessed using the HNAT in Site A, but none were offered this option during the tape-recorded interactions. This may be an artefact of the research, in that health visitors may have felt it inappropriate to record a conversation where the client refused, but there was no mention or discussion of how this might be negotiated with clients during the data collection period for this or the wider study from which the interactions were recorded. In contrast, in site B, only one health visitor referred to the assessment process (as in Extract 12). Since the other clients were not informed that an assessment process was underway, they had no opportunity to indicate their own view of their needs; instead, the final decision about the clients’ level of need and the services to be offered remained, exclusively and covertly, with the health visitor.

The final phase of the interaction centred upon the planning for future interventions and closure of the visit. A striking feature of the interactions, in both sites, was that almost without exception the health visitor gave the clients no opportunity to ask questions until the end of the encounter when it was evident that the business of the visit (that is, the assessment of need) was concluded. In Site B, the visits ended, at this point, with the health visitor indicating to the client that she had achieved her objectives. Arrangements for future contact were made with an explanation of the kinds of needs that would justify this, as in Extract 13:

Extract 13 HV/C3B

Line 455 HV and often we can help out you know if its to do with their welfare

Line 456 and their well being you know=

Line 457 Client yeah

Line 458 HV keeping them fit

Line 459 Just wait and see when this baby arrives if you’re fraught or

Line 460 whatever just let us know you've got the number haven't you

Line 461 //and we’ll see what we can do

Line 462 Client //yeah

Line 463 HV right I think that's it now I think we’re finished

In Site A, where the form-filling was more overt, it was more common for the health visitors to signal that, now the assessment was complete, a more informal and relaxed conversation could take place whilst concluding the whole business of the visit. The frequency of home visits was invariably determined by the health visitor with clients in both sites being encouraged to attend the baby clinic as a means of accessing the health visiting service.

The process overall

The needs assessment process in these two sites, each using a different structured needs assessment tool, was typified by two things: first, a lack of client participation in identifying and determining their health needs and second, a process of informing the client about an ‘appropriate’ use of the service, which was characterised by seeking help for problems that were normatively defined. The interaction in Site A was largely structured by the questionnaire format, resulting in a uniformity that is more commonly seen in medical history taking, as reported in ten Have (1991), than in the private conversations to which earlier studies likened health visitor/client interactions (Dingwall and Robinson 1990). During the assessment phase of the interaction the health visitors systematically asked closed questions and the client answered with short responses.

This element of regularity and control was much less obvious in Site B where the questions covered similar topics but were more open and exploratory in nature. Nevertheless, the health visitor decided the agenda and the topics for discussion, which is similar to the patterns in other CA studies of health visiting (Heritage and Sefi 1992, Kendall 1993). In most instances in Site B, the visit was conducted without the clients’ prior knowledge that her needs were being assessed or that the level of future health visiting service would be determined by the outcome of that assessment. The risk of child abuse, which had been a driving force in developing and implementing the early intervention strategy, was not mentioned to clients.

In both sites, the health visitors departed from the ‘official’ versions of how the assessment tools were to be used in different ways. Despite this variation, in both sites the assessment process was intrusive in many ways and either overtly or covertly controlling. The health visitors adopted a stance that ensured control of the agenda and offered only clearly-bounded opportunities for the client to contribute to the specification of their own needs. During the assessment process the health visitors often actively disattended to the cues tentatively offered by their clients. The health visitor controlled the interaction during the information/advice-giving sequences, and during closure and the determination of future contact. The purpose of the needs assessment visit was revealed as one concerned with identifying problems from within a pre-determined list and informing clients about the nature of approved needs for which services may be used and how they can be accessed. Thus, the organisational requirement to retain a normative focus of the health visiting service was secured and service boundaries were maintained.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

As in earlier analyses of interactions between health visitors and their clients (Heritage and Sefi 1992, Kendall 1993), this study revealed conversations characterised by a non-participative style, with the agenda being set by the health visitors and clients acquiescing in that approach. Elements of resistance, highlighted elsewhere (Bloor and McIntosh 1990, Peckover 2002) recurred in this study, with few clients expanding on the topics offered to them or introducing new issues when, their official business finally concluded, the health visitors invited questions from them.

Two aspects of this study appear new and deserving of further consideration. First, there seems to be a marked change from the earlier apparently ‘agenda-free’ conversations noted by Dingwall and Robinson (1990). The three-stage process of needs assessment identified in this study (introducing the instrument, questioning process, then outcome of assessment), is significantly different from health visiting processes documented in the professional literature (Cowley 1991, 1995, Chalmers 1992, Chalmers and Luker 1992). Taken together with the resemblance between the structured conversations and the doctor/patient interactions reviewed by ten Have, this suggests a step-change in the medicalisation of health visiting, where structured HNATs are used. This may help to explain the second aspect: the explicit resistance strategies of health visitors towards using the HNATs in the manner required of them by their employers. In Site A, the overt use of the HNAT as a questionnaire directly opposed instructions from managers that such a ‘checklist approach’ should be avoided. In Site B, the covert embedding of the questions enabled health visitors to continue what may be considered a more ‘traditional approach’ to practice.

The need to legitimise their presence with clients is not, in general, a problem faced by nurses or doctors, since patients usually initiate the contact. By contrast, the unsolicited and proactive nature of health visiting work requires that the legitimacy of the interaction be established before any meaningful work can take place (Cowley 1991). The establishment of a non-judgemental and listening relationship, which allows flexibility of purpose, along with giving, receiving and trust on both sides, appears central in achieving this (Chalmers and Luker 1991, Chalmers 1992, Cowley 1991, 1995, de la Cuesta 1994, Peckover 2002). The loose conversational forms noted by Dingwall and Robinson (1990), Heritage (2002), and to a great extent in site B in this study, show how these processes are manifested.

On the other hand, the assessment proforma used by the health visitors in this study appear to set up conversations that mirror those carried out in the medical interactions reviewed by ten Have (1991). That is, they contain a series of questions for which the health visitor rarely provides motivation or explanation; topic changes are abrupt and the patient/client experience is selectively ignored before a decision about the level of need, analogous to the ‘diagnosis’, is reached by the health visitor and agreed by the client. The key difference between medical encounters and health visiting ones, that the professional rather than the client initiates the contact, is not accounted for within this form of sequence and there is no apparent mechanism for legitimising the contact or the assessment.

The apparent resistance of the health visitors in the two sites may be considered in the context of Foucault's description of pastoral power, in which caring and support are offered by professionals, so that they can gain access to the most personal and intimate aspects of people's lives, for purposes of surveillance and control (Peckover 2002). This translates, from the health visiting perspective, to a trade-off in which services and support are provided on terms decided by the client, to allow a trust relationship to develop. It is only when that has happened, as their professional literature testifies, that meaningful health promotion and support can take place (Chalmers and Luker 1991, Chalmers 1992, de la Cuesta 1994). Offering an open, general interest and caring approach enables the relationship to develop and allows health visitors to provide support and help even if there is no specific diagnostic label (Cowley 1995). The structured assessment tools impede the relationship-building process and provision and suggest disapproval of such non-specific support; instead, they are used to identify an organisationally-approved diagnostic label, without which health visiting services are disallowed.

Effectively, the mechanisms used to acquire the information for the structured assessments followed two different formats, neither of which are wholly satisfactory for health visiting practice or clients receiving the service. By weaving the structured approach to health needs assessment into their everyday practice, the health visitors in Site B succeeded in complying with the new, more explicit surveillance requirements. In doing so, they effectively deceived the clients about the purpose of their questions, which originate from a child protection agenda and serve (at least in part) to maintain a boundary around the health visiting service, ensuring that only those with ‘approved needs’ are targeted by being offered follow up services. Since, in most instances, they did not inform clients that their personal details were to be so documented, the health visitors exercised absolute power over the label to be afforded to the client. Thus, anyone deemed in need of a service, because they reveal enough risk factors to reach an epidemiological threshold for child abuse or family dysfunction, would be neither aware that they had been so labelled nor given the opportunity to change it. Alternatively, someone wanting help and support may be unable to access it, again, without knowing the reason why they are denied the service. At the same time as obscuring the purpose to the clients, embedding the questions in their conversations serves to restrict organisational control over the health visitors’ practice.

In Site A, the health visitors were faced with a clearer requirement to return the results of their assessments to their managers for scrutiny; they adapted their practice in different ways. First, they displayed some resistance strategies, evidenced by their discomfort in questioning and distancing themselves from the questionnaire by attributing it to their employing Trust. The use of the needs assessment proforma was clearly separate from the rest of their practice, being administered as a bureaucratic, form-filling exercise, an approach also identified by Heritage (2002). Not only was the start of the process signalled explicitly, but once it was complete, the health visitors resumed ‘normal practice’, reverting to their more usual, loosely formulated conversational style. The clear identification of the form-filling as symbolically different from the rest of the visit may signal to clients that they need not take any labels too seriously. The determined style of questioning, however, was so overtly controlling and, at times, insensitive, that it seems unlikely that any client subject to the process would feel able to disregard it.

Ostensibly, the HNATs are claimed as feeding into the ‘public health role’ of health visitors, which is concerned with populations rather than with individual families (Department of Health 2001). It should be noted, in this connection, that in neither site in this study were the results of the needs assessments used for epidemiological purposes, but the medicalising impact of the HNATs on health visiting practice is clear. Neither bureaucratic form-filling nor covert eliciting of information are satisfactory approaches, either in practical or ethical terms, from the point of view of health visitors or clients. Both approaches did allow the health visitors to achieve the organisational requirements for completion of the HNAT. Either separating the procedure from the rest of their practice or hiding it from the client would, in effect, allow the health visitors to revert to an individualised, loosely formulated approach to practice once the information-gathering exercise was concluded. It might be anticipated that, should the deception be revealed, or the bureaucracy considered offensive by clients (as indicated in the wider study in Site A, see Houston and Cowley 2003), the task of rebuilding relationships would be rendered extremely difficult and the legitimacy of health visiting exposed to further question. Further, those whose self-defined needs differ from the normative definitions would not be included, since services are targeted according to the population-wide indicators elicited through the HNAT, instead of individually-defined needs.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Through much of the last quarter of the 20th century, social welfare and family concerns were low on the political agenda and health visiting struggled, as a profession, to maintain legitimacy and purpose. A shift in focus towards public and population health has not diminished health visiting interest in mothers and young children, since experience in the early years is so significant for later health (Independent Inquiry into Inequalities in Health 1998, Wadsworth 1999, Shonkoff and Phillips 2000). The need to identify vulnerable individuals remains high and, despite the official distancing of health visitors from the child protection agenda, it is noteworthy that the majority of guidelines intended for their use in prioritising continue to focus on risk factors related to family dysfunction and abuse. The emphasis on a population-wide approach, however, has challenged the former loosely-specified approaches to health visiting practice that represented ‘in effect, the systematic ethnographic study of a community by an expert in public health’ (Dingwall and Robinson 1990: 268).

Structured and pre-defined needs-assessment schedules, of the kind featured in this study, have been implemented in many places. They suggest that health visiting activities will be sponsored as legitimate only if they conform to medically-defined, population-approaches to public health; clients’ needs can be serviced only if they reach the threshold of epidemiologically-defined risk. In this respect, the HNATs serve as a marker in the medicalisation of the health visiting profession and a step away from their former role in providing general social surveillance, support and prevention. This study has shown that the health visitors were able to achieve the organisational requirements to complete the assessment schedules, at the expense of the flexibility needed to legitimise contact with individual clients. It is not clear, in this new occupational formulation, whether health visiting will be able to maintain the universal contact with families that enables support to reach some of the most vulnerable and isolated members of society. Without this grass-roots legitimacy, it seems unlikely that the changed role will be sustainable in the long term.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Thanks are due to the Florence Nightingale School of Nursing and Midwifery, King's College London whose special research funds paid for data collection from site A. Thanks are also due to the research participants, including the clients and the practitioners who exposed their practice to scrutiny and the Trust managers who allowed us completely open access to their organisations. We much appreciate the invaluable guidance and suggestions from two anonymous reviewers and the editorial team, which helped us to refine the manuscript and understand our data more clearly. Interpretation of the results of the study, as always, lies with the researchers.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Background to the study
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References
  • Abbott, P. and Sapsford, R. (1990) Health visiting: policing the family? In Abbott, P. and Wallace, C. (eds) The Sociology of the Caring Professions. Basingstoke: The Falmer Press.
  • Appleton, J. (1997) Establishing the validity and reliability of clinical practice guidelines used to identify families requiring increased health visitor support, Public Health, 111, 10713.
  • Appleton, J. and Cowley, S. (2003) Valuing professional judgement in health visiting practice, Community Practitioner, 76, 21520.
  • Audit Commission (1994) Seen But Not Heard. Co-ordinating Community Child Health and Social Services for Children in Need. London: HMSO.
  • Barker, W. (1990) Practical and ethical doubts about screening for child abuse, Health Visitor, 63, 1417.
  • Barker, W. (1996) Form filling: a demotion of professional skills, Community Practitioner, 69, 4078.
  • Bidmead, C., Davis, H. and Day, C. (2002) Partnership working: what does it really mean? Community Practitioner, 75, 2569.
  • Bloor, M. and McIntosh, J. (1990) Surveillance and concealment: a comparison of techniques of client resistance in therapeutic communities and health visiting. In Burley, S. and McKegany, N. (eds) Readings in Medical Sociology. London: Routledge.
  • Browne, K. (1995) Preventing child maltreatment through community nursing, Journal of Advanced Nursing, 23, 499507.
  • Chalmers, K. (1993) Searching for health needs: the work of health visiting, Journal of Advanced Nursing, 18, 90011.
  • Chalmers, K. and Luker, K. (1991) The development of the health visitor–client relationship, Scandinavian Journal of Caring Sciences, 5, 3341.
  • Chalmers, K. (1992) Giving and receiving: an empirically derived theory on health visiting practice, Journal of Advanced Nursing, 17, 131725.
  • Cowley, S. (1991) A symbolic awareness context identified through a grounded theory of health visiting, Journal of Advanced Nursing, 16, 64856.
  • Cowley, S. (1995) In health visiting, the routine visit is one that has passed, Journal of Advanced Nursing, 22, 27684.
  • Cowley, S. and Houston, A. (2003) A structured health needs assessment tool: acceptability and effectiveness for health visiting, Journal of Advanced Nursing, 43, 110.
  • Crompton, K., Davies, M. and Humphris, A. (1998) Child protection: developing an early intervention strategy, Health Visitor, 71, 568.
  • Council for Education and Training of Health Visitors (1977) An Investigation into the Principles and Practice of Health Visiting. London: Council for Education and Training of Health Visitors.
  • Davies, C. (1988) The health visitor as mothers friend: a woman's place in public health 1900–1914, Social History of Medicine, 1, 3959.
  • Department of Health (2001) Health Visitor Practice Development Resource Pack. London: Department of Health.
  • De La Cuesta, C. (1994) Relationships in health visiting: enabling and mediating, International Journal of Nursing Studies, 31, 4519.
  • Dingwall, R. (1977) Collectivism, regionalism and feminism: health visiting and British social policy 1850–1975, Journal of Social Policy, 6, 291315.
  • Dingwall, R. (1982) Community nursing and civil liberty, Journal of Advanced Nursing, 7, 33746.
  • Dingwall, R., Rafferty, A.M. and Webster, C. (1988) An Introduction to the Social History of Nursing. London: Routledge.
  • Dingwall, R. and Robinson, K. (1990) Policing the family? Health Visiting and the public surveillance of private behaviour. In Gubrium, J. and Sankar, A. (eds) The Home Care Experience: Ethnography and Policy. Newbury Park, CA: Sage.
  • Foucault, M. (1976) The Birth of the Clinic: an Archaeology of Medical Perceptions. London: Tavistock.
  • Foucault, M. (1979) Discipline and Punish: the Birth of the Prison. London: Penguin Books.
  • Foucault, M. (1984) The History of Sexuality, Volume 1. London: Penguin Books.
  • Gill, R. (2000) Discourse analysis. In Bauer, M. and Gaskell, G. (eds) Qualitative Researching with Text, Image and Sound: a Practical Handbook. London: Sage.
  • Goddard, C.R., Saunders, B.J. and Stanley, J.R. (1999) Structured risk assessment procedures: instruments of abuse? Child Abuse Review, 8, 25163.
  • Heritage, J. (1997) Conversation analysis and institutional talk. In Silverman, D. (ed.) Qualitative Research; Theory, Method and Practice. London: Sage Publications Ltd.
  • Heritage, J. (2002) Ad hoc inquiries: two preferences in the design of routine questions in an open context. In Maynard, D.W., Houtkoop-Steenstra, H., Schaeffer, N.C. and Van Der Zouwen, J. (eds) Standardization and Tacit Knowledge: Interaction and Practice in the Survey Interview. New York: John Wiley and Sons Inc.
  • Heritage, J. and Lindström, A. (1998) Motherhood, medicine and morality: scenes from a medical encounter, Research on Language and Social Interaction, 31, 3 and 4, 397438.
  • Heritage, J. and Sefi, S. (1992) Dilemmas of advice: aspects of the delivery and reception of advice in interactions between health visitors and first-time mothers. In Drew, P. and Heritage, J. (eds) Talk at Work: Interaction in Institutional Settings. Studies in Interactional Sociolinguistics 8. Cambridge: Cambridge University Press.
  • Houston, A. and Cowley, S. (2002) An empowerment approach to needs assessment in health visiting practice, Journal of Clinical Nursing, 11, 64050.
  • Independent Inquiry into Inequalities in Health (1998) Independent Inquiry into Inequalities in Health (chair, Acheson, D.) London: The Stationery Office.
  • Kendall, S. (1993) Do health visitors promote client participation? An analysis of the health visitor-client interaction, Journal of Clinical Nursing, 2, 1039.
  • Machen, I. (1996) The relevance of health visiting policy to contemporary mothers, Journal of Advanced Nursing, 24, 3506.
  • Miller, G. and Silverman, D. (1995) Troubles talk and counselling discourse: a comparative study, The Sociological Quarterly, 36, 72447.
  • Mitcheson, J. and Cowley, S. (2003) Empowerment or control? An analysis of the extent to which client participation is enabled during health visitor/client interactions using a structured health needs assessment tool, International Journal of Nursing Studies, 40, 41326.
  • Naughton, A. and Heath, A. (2001) Developing an early intervention programme to prevent child maltreatment, Child Abuse Review, 10, 8596.
  • NHS Executive (1996) Child Health in the Community: a Guide to Good Practice. London: Department of Health.
  • Normandale, S. (2001) A study of mothers’ perceptions of the health visiting role, Community Practitioner, 74, 146150.
  • Peckover, S. (2002) Supporting and policing mothers: an analysis of the disciplinary practices of health visiting, Journal of Advanced Nursing, 38, 36977.
  • Perakyla, A. and Silverman, D. (1991) Reinterpreting speech-exchange systems: communication formats in AIDS counselling, Sociology, 25, 62751.
  • Psathas, G. and Anderson, T. (1990) The ‘practice’ of transcription in conversation analysis, Semiotica, 78, 7599.
  • Roberts, C. (1996) The proof of the pudding, Health Service Journal, 106, 5494, 27.
  • Robinson, K.M. (1986) The social construction of health visiting. Unpublished PhD thesis, London: Polytechnic of the South Bank.
  • Sacks, H., Schegloff, E.A. and Jefferson, G. (1974) A simplest systematics for the organization of turn taking for conversation, Language, 50, 696735.
  • Sacks, H. (1992) Lectures on Conversation. Volume 1. Oxford: Blackwell.
  • Sefi, S. (1985) The First Visit – a study of health visitor-mother verbal interaction. Unpublished MA Dissertation. Coventry: University of Warwick.
  • Shonkoff, J. and Phillips, D. (eds) (2000) From Neurons to Neighbourhoods: the Science of Early Childhood Development. Washington: National Academy Press.
  • Ten Have, P. (1991) Talk and institution: a reconsideration of the ‘assymetry’ of doctor-patient interaction. In Boden, D. and Zimmerman, D.H. (eds) Talk and Social Structure. Cambridge: Polity.
  • Ten Have, P. (1999) Doing Conversation Analysis: a Practical Guide. London: Sage.
  • Wadsworth, M. (1999) Early life. In Marmot, M. and Wilkinson, R. (eds) Social Determinants of Health. Oxford: Oxford University Press.