OUP user menu

Results of a national asthma campaign survey of primary care in Scotland

Gaylor Hoskins, Colin McCowan, Peter T. Donnan, James A. R. Friend, Liesl M. Osman
DOI: http://dx.doi.org/10.1093/intqhc/mzi036 209-215 First published online: 14 April 2005


Objectives. To identify within primary care in Scotland how far procedures for asthma review and patient education match guideline recommendations.

Design and setting. Telephone survey of a one in four stratified random sample of all 1058 general practices in Scotland.

Participants. Practice nurses, general practitioners.

Main outcome measures. Number of practices matching guideline recommendations for asthma review, targeting of care, use of structured asthma records, provision of management plans, education, and regular audit.

Results. Of 276 general practices contacted 91% (251) completed the questionnaire; 93% (228) ran an asthma review service; 74% (166) employed a specially trained asthma nurse; 39% (106) had a policy for providing action plans; 63% (155) had carried out an asthma audit in the previous 3 years; 76% (218) used a structured tool in consultations, 46% with use of computer technology, 34% used only a manual stamp. Sixty-six per cent (173) had searched for patients overusing β2 agonists; 32% (79) had searched for patients on medication treatment step 3 and above. Single- or two-partner practices were less likely to follow guideline recommendations but neither rurality nor deprivation was related to guideline compliance.

Conclusions. Three-quarters of Scottish general practices have trained asthma nurses and offer patients asthma review, but only a minority have proactive care procedures for targeting patients or a policy for providing patients with action plans. Practice systems are underused for identifying ‘at-risk’ patients. There is a need for proactive procedures and provision of self-management materials to patients. Access to trained asthma nurses needs to be improved.

  • asthma
  • clinical
  • guidelines
  • primary care
  • process of care
  • review
  • structured care

It has been commented that guidelines are a tool for closing the gap between what clinicians do, and what the scientific evidence supports. There are barriers to the full implementation of guidelines. Providing information alone is ineffective. Inadequate time, resources, and support are identified by health professionals in primary care as factors that inhibit change [1]. Health professionals are most likely to put guidelines into practice when they are developed in a multidisciplinary approach with clear presentation of levels of supporting evidence [2] and where they are developed by those who implement them, and thus ‘owned’ by participants [3]. The Scottish Intercollegiate Guidelines Network (SIGN) has developed such a methodology, and it has been applied in the development of more than 50 evidence-based sets of clinical guidelines [4]. Since 1993 the British Thoracic Society asthma guidelines have been developed in conjunction with the SIGN (known as the SIGN/BTS guidelines). The most recent revision was published in 2003 [5].

The management of asthma is a major task within general practice. In a recent study of 319 Scottish general practices, 43% of patients with asthma attended at least once over a 12 month period for an urgent consultation with their general practitioner (GP) [6].

Although reviews of hospital care have been carried out [7,8] only a little published data exists of how far non-acute management of asthma in general practice matches existing guidelines. For instance the SIGN/BTS guidelines recommend, based on strong evidence [911] that structured care is enhanced if carried out by a nurse with specialist training in asthma. One regional study found that among nurses carrying out asthma reviews, 56% did not have specialist asthma training [11]. Another study involving one-third of all Scottish practices found that only 27% of practices used a structured review tool for recording consultation data [12]. We do not know how many practices have put in place an asthma care audit loop. We know little about practice provision of action plans to patients although this has been recommended for several years by the guidelines.

In asthma, evidence is strong that patient education and self-management are important in successful clinical management, and these areas are emphasized in asthma guidelines. Asthma UK Scotland was funded by the Scottish Executive to carry out a baseline survey of how far existing practice matched guideline recommendations in the period immediately before publication of revised guidelines. The guideline areas of organization and delivery of care and patient education and self-management were the focus of the study and this paper reports the result of the survey (Table 1).

View this table:
Table 1

Number of practices following guideline recommendations [4–5]

ProvisionNumber of practices (%)1
Access to reviewReview service provided228 (93)
Open access appointment system186 (84)
Fixed time clinic only34 (12)
Asthma register233 (95)
Register updated annually188 (76)
Asthma trained nurse166 (74)
Doctor asthma trained/interest group32 (11)
Structured recordingAny structured tool for recording218 (90)
Manual stamp/paper pro forma80 (30)
Computer template76 (27)
Manual stamp and computer template21 (10)
Free text only40 (21)
Action planPractice policy for providing plans to patients176 (39)
Supply of action plans available179 (75)
System searches for at-risk patients at least annuallyOn any criterion178 (68)
Patients on BTS step 3+5 (2)
Frequent users of β2 agonists92 (33)
BTS step 3+ and frequent β2 agonist users55 (23)
Frequent β2 agonist users and URTI in children7 (2)
BTS step 3+ and frequent β2 agonist users and URTI19 (8)
Never73 (32)
Searches for acute asthma attackSpecific search carried out59 (28)
No specific search but continually monitored manually89 (38)
AuditAnnual44 (20)
Within the last 3 years155 (63)
Internal93 (45)
External68 (27)
  • URTI, upper respiratory tract infection.

  • 1 All percentages are adjusted to give results representative of Scotland overall.


Sampling method

Details of all practices in Scotland (n = 1058) were obtained from the Primary Care Information Group of the Information and Statistics Division (ISD) of the National Health Service (NHS) in Scotland. Information on each practice included age, sex, and social deprivation of the practice population, practice code, number of whole-time equivalent (WTE) partners, and indicators of rurality. ISD categorized practices as rural where a third or more of the practice patients lived over 3 miles from their central general practice surgery and hence attracted rural practice payments [13,14]. The practice sample was stratified by NHS board, and partners in the practice (1, 2–4, and 5+) to ensure capture of information from a wide geographical area and representation of practices of all sizes. Selection of practices within each stratum was achieved by using random number generation. A sample size of 250 was chosen to give sufficient precision (± 5%) for prevalence values. A different sampling fraction for each cell was chosen because uniform sampling would have resulted in only one or two from the very small NHS boards, such as Orkney and Shetland. This method involved over-sampling in rare strata (e.g. Western Isles) and under-sampling in more common areas (e.g. Lothian). The percentages obtained as responses to the questionnaire were weighted by the stratum-specific weight according to the whole of Scotland. Hence, the weighted percentage adjusts for the sampling, to give results which are representative of the whole of Scotland.

Survey method

The project advisory group consisted of representatives of Asthma UK, medical, nursing, and pharmacy professionals, academic researchers in primary care, and patient representatives.

The most recent previous asthma guidelines had been published in 1998 [15]. The study advisory group was provided with a pre-publication copy of the 2003 guidelines. Both the 1998 and 2003 guidelines placed emphasis on the areas covered by the survey: targeting of review in primary care and patient education and provision of action plans, need for trained asthma nurses, asthma registers, specialist doctor training, and audit tools.

The 2003 guidelines [5] did not remove any of the recommendations of the 1998 guidelines, but they now gave specific criteria for identifying patients for targeted review, which had not been done in the 1998 guidelines. These criteria were used in the survey.

The eight areas of recommendation for organization of care and patient education and self-management identified within the 1998 guidelines were also part of the 2003 recommendations. It was agreed that these were suitable for assessment through a brief structured survey.

  1. Asthma register: Did the practice maintain a list of people with asthma?

  2. Review service: Were people with asthma reviewed regularly?

  3. Trained asthma nurse: Was review carried out by a nurse with training in asthma management?

  4. Assessment tool: Was clinical review structured, utilizing a standard recording system?

  5. Specialist doctor training: Did doctors in the practice have Continuing Medical Education training in asthma management?

  6. Regular audit: Was it carried out?

  7. Action plan: Were customized asthma action plans offered to all people with asthma?

  8. Targeting of care: Was care targeted to patients at risk?

    1. Children over 5 years with frequent consultations with respiratory symptoms.

    2. Patients using large quantities of β2 agonists.

    3. Patients at Step 3 or above.

    4. Patients with frequent oral steroid courses, Accident and Emergency attendances, or hospital admissions.

The group discussed the problem of reliability of self-reporting, particularly of areas 7 and 8. It was decided that although areas 1–6 could be fairly reliably assessed by simple Yes/No questions, recommendations 7 and 8 were more difficult to assess. Questions asking whether the practice gave action plans to all patients, and whether the practice targeted care to at-risk patients were replaced with more specific questions on objective actions taken by practices which would be prerequisites for recommendations 7 and 8. These were:

  1. To ask practices whether they had a written protocol for the provision of management plans, rather than ask if they gave action plans to all patients with asthma.

  2. To ask practices if they regularly used their practice systems to search for patients who fitted the ‘at-risk’ criteria described above.

A structured telephone interview questionnaire was developed and trialled. The interview used a script introducing the researcher, giving an explanation of what the study was about and on whose behalf it was being carried out [16]. Tick box options were used for speed of recording of responses.

The survey was carried out between July and September 2002. The Research Committee for Medical Ethics, Edinburgh, was informed of the survey but did not require a formal ethics application to be submitted.

Statistical methods

All analyses were carried out using the Statistical Package for Social Sciences (SPSS) v. 10. The results are presented as a raw number with a weighted percentage which gives the response adjusted to give results which are representative of the whole of Scotland. Tests were carried out in univariate analyses to assess the importance of deprivation (dichotomized into Carstairs categories 1–4 Not Deprived versus 5–7 Deprived [17]); rurality (defined by practices in receipt of rural practice payments [13,14]); and partner size (dichotomized to one or two partners versus three or more partners) on binary outcomes. A multiple logistic regression model was used to assess associations of important factors. Variables with at least univariate associations with P < 0.2 were considered for potential inclusion in the regression [18]. Selection of variables for the final model was at the conventional 5% significance level using forward selection. Results are reported as odds ratios and 95% confidence intervals (CIs).


Responses were obtained from 251 of the 276 general practices contacted (Figure 1). For 25 practices it proved impossible to organize an agreed time for the telephone interview, or faxed questionnaires were not returned. Most questionnaires were completed by a practice nurse [202 (80%)]; 42 (17%) were completed by a doctor, and 7 (3%) were completed by the practice manager.

Table 1 shows that 90% or more of practices reported having an asthma register, providing a review service, and using a structured tool for recording. Between 63% and 76% of practices updated their register annually, carried out regular searches for at-risk patients, had an asthma trained nurse and had carried out audit within the previous 3 years. Less than 40% of practices had a practice policy for providing action plans, carried out searches for patients who had had acute attacks, or had carried out audit within the last year.

Practice size

Practices with only one or two partners were less likely to run an asthma review service, have an asthma trained nurse, use a review assessment tool, and have a written practice self-management plan protocol (Table 2).

View this table:
Table 2

Practice characteristics and conformity to Guideline recommendations

No. of practices sampled (%)Review service, n (%)Trained nurse, n (%)Register, n (%)Assessment tool, n (%)Systems search, n (%)Audit in last 3 years, n (%)SMP written protocol, n (%)
Scotland251228 (93)166 (74)233 (95)218 (90)178 (68)155 (63)106 (39)
Practice rural126 (10)49 (89)31 (60)50 (88.5)51 (88.5)45 (77)35 (64)26 (42)
Urban225 (90)179 (94)135 (76)183 (96)167 (90)133 (67)120 (63)80 (39)
P value0.540.360.221.000.410.900.81
Partner size
    Small practice (1–2 partners)89 (35)85 (88)47 (58)83 (94)79 (82)64 (64)58 (62)37 (29)
    Large practice (3+ partners)162 (65)143 (96)119 (83)145 (96)139 (94)114 (70)97 (64)69 (45)
P value0.020.010.860.010.340.780.03
Health Board deprivation: Carstairs
    Greater than Scottish average128 (51)73 (96)63 (81)74 (98)72 (94)49 (62)51 (66)35 (37)
    Less than Scottish average123 (49)155 (90)103 (67)159 (92)146 (85)129 (75)104 (59)71 (42)
P value0.
  • 1 Practice rurality as defined by the ISD.

  • 2 All percentages are adjusted to give results representative of Scotland overall.

Social deprivation and guideline compliance

Having an asthma register and using an assessment tool were more frequent in practices with higher deprivation. Systematic searches were more frequent in the least deprived practices (Table 2).

Trained asthma nurse

The presence of a trained practice asthma nurse is likely to be an important element in the provision of asthma care. Consequently, logistic regression analysis assessed this outcome in relation to individual practice rurality, practice partner size, deprivation as measured by the Carstairs index, and the presence of an asthma interested GP in the practice. A greater number of partners in the practice and greater deprivation in the practice area are independently associated with the presence of a practice nurse (Table 3).

View this table:
Table 3

Relationship between the presence of a trained practice asthma nurse and other variables

Unadjusted logistic regression analysisAdjusted logistic regression analysis
Odds ratio95% CIP valueOdds ratio95% CIP value
Practice rurality0.710.32−1.610.411.220.49−3.0150.67
Number of WTE partners1.351.09−1.680.011.381.10−1.7440.01
Deprivation by Carstairs2.180.97−4.860.062.721.12−6.650.03
Asthma interested GP1.150.57−2.300.700.9950.48−2.070.99


The findings of this survey highlight the gaps in service provision for asthma care. Although almost all practices offered some level of review service, this was not always given by a nurse with specialist asthma training. In many practices asthma registers were not being regularly updated, few utilized their computer systems to carry out proactive searches for patients in at-risk groups; and only a third of practices had an agreed policy for providing action plans to patients.

The survey results suggest that there has been improvement since the 1998 guideline publication. Seventy-four per cent of practices had a specialist asthma trained nurse, which is almost double the 36% of practices having a trained nurse in one major Scottish region in 1996 [19].

Ninety per cent of practices purported to be using some form of structured record with 68% using a distinct tool to record the data, compared with the 27% found in a previous Scottish sample [12]. Use of a tool, whether manual or computer based, gives the assurance that data which are considered the minimum required for assessing a patient’s condition is collected. In almost half the practices surveyed computer technology was employed as the tool of choice. However, a sizeable minority of practices (21%), although basing the consultation around the Royal College of Physicians three questions for asthma [20], record the answers as free text in notes. Most of the practices using this method considered this to be a ‘structured’ review. It is not possible within the scope of this type of survey to determine whether sufficient information was being asked or recorded using this format. What is clear is that practices need to be supported and encouraged to use more formal tools for carrying out and recording asthma review consultations.

Targeting of care to those most in need is a more effective use of resources. Although most general practices had a ‘register’ of patients with asthma many did not update it regularly. This restricts the ability of the practice to identify patients most in need of review. The provision of specific criteria for identifying patients at risk has been a key addition to the recommendation for regular review in the guidelines. The use of practice computer systems to search for ‘at-risk’ patients, proposed as a key facet of asthma management in the 2003 revised guidelines, was not a widespread practice. Where this was utilized it was likely to identify overuse of a bronchodilator drug. This is a criterion for targeting patients with loss of asthma control but patients suffering symptoms do not always overuse their bronchodilator [21]. Thus, the forthcoming 2003 guideline recommendation to search for other categories of ‘at-risk’ patients was not being carried out by most of the practices at the time of the survey in late 2002.

As has been found previously [2,22] it was clear from the survey that the majority of the educational material used within practices was sourced from the pharmaceutical industry. Although this material is largely accurate and up to date, terminology is not consistent between different pharmaceutical companies, and emphases differ [22]. This can be confusing for patients. There should perhaps be less reliance on commercial provision of what is after all a core responsibility.

The 1998 and 2003 guidelines give a grade A recommendation that all patients and particularly those who have been in hospital should be offered an asthma management plan. We were unable to assess directly the provision of action plans to patients but protocols for the use of plans had not been agreed within almost two-thirds of practices. Although practices stated that self-management plan materials were widely available, the lack of protocols suggests that practices do not have clear guidelines for their staff on use of action plans. It further suggests that practices are not aware of the need for training staff in the provision of action plans and of how to actively encourage patients to use action plans effectively. Evidence of protocols within practices would be evidence that patients get a consistent agreed message from all members of the health care team within a practice and that a single type of plan is agreed upon. It would appear that action plan provision is still at a low level within Scottish primary care.

The most significant correlate of compliance with guideline recommendations was the size of the practice. It has been found previously that asthma admission rates are strongly associated with size of practice partnership, with one- or two-partner practices having almost double the admission rates of larger practices [23]. In the present study smaller practices were less likely to have an asthma trained nurse, to use assessment tools, or to have a protocol for the use of action plans.

It would appear that the lack of an asthma nurse has repercussions for the ability of a practice to create the best recommended environment for care of patients with asthma. Smaller practices with fewer resources are at a disadvantage in implementing recommended organization of asthma care.

Methodological issues

Telephone interviewing enabled the researcher to gather data quickly, and over a large geographical area. A structured format entailed that the interviewer asked exactly the same questions of each interviewee, but there was also flexibility for the interviewee to clarify the meaning of questions, negotiate or elaborate on responses, and make comments. As anticipated, this method produced a high response rate. The drawback in utilizing this method of data collection is that it was not possible to validate the information provided. Practices may have overestimated the range and quality of the structured care they provided.

The randomization procedures guaranteed that practices of all sizes and from all parts of Scotland were represented. Weighting each practice before randomization ensured that the results were indicative of the Scottish picture.

The survey was designed to ascertain the level of provision within primary care to provide an adequate level of service to people with asthma. As a result no data were collected on patient outcome. We cannot assume that where recommended practice is not being achieved poor patient outcome is inevitable. This may be particularly so in small rural practices.

The survey was conducted by an independent research unit on behalf of Asthma UK Scotland and supported by a grant from the Scottish Executive Health Department. Participating practices were made aware of the fact that Asthma UK Scotland wanted to ascertain the level of activity for managing asthma to enable them to highlight the requirements for improving standards. If practices perceived this to be an ‘official review’ there is a danger that practices may have tended to enhance the amount of activity they were involved in. However, as it was made clear that the survey was to be used as a tool for lobbying for extra resources and support for primary care, overstating their activity would have been counterproductive.


This survey has provided a baseline measurement of compliance to guideline recommendations on organization of care in Scottish primary care practices. It suggests that practice procedures to address guideline recommendations have been developed in the areas of specialist training for nurses, and practice data collection systems. These improvements may have arisen due to organizational changes in primary care, such as fund-holding and the introduction of nurse practitioners over the 5 years between 1998 and 2003, or a greater awareness of asthma and its management caused by the higher profile of professional bodies such as the General Practice Airways Group and charities such as Asthma UK. At present, the biggest gap appears to be in proactive targeting of patients, and proactive assistance to patients to develop action plans. The survey suggests that the methodology developed by SIGN is achieving success in dissemination and take-up of guideline recommendations, and has developed strategies to assist health professionals attain the good practice recommended by guidelines. However, incentives and assistance to structure care as recommended may be necessary in smaller practices.

The survey suggests that practices need support in developing procedures to provide action plans to at-risk patients, and need instruction on how best to record data and to interrogate the record system to identify active asthmatics and those at risk.


To all the GPs, practice nurses, and practice managers who contributed to the findings of this survey, thank you. The project was funded by a grant from the Scottish Executive Health Department to Asthma UK Scotland and was a collaborative venture between the charity, invited lay and professional members of the Steering Group, the Asthma Research Unit, Tayside Centre for General Practice, the University of Dundee, and the University of Aberdeen.


View Abstract