Evidence in action—guidelines directing workload
- 1Current Care, Finnish Medical Society Duodecim, Helsinki, Finland
- 2Department of Public Health, University of Helsinki, Helsinki, Finland
- Correspondence to Raija Sipilä, Finnish Medical Society Duodecim, PO Box 713, Kalevankatu 3 B, Helsinki 00101, Finland;
Contributors EK, TT and RS designed the audit process. RS, MK and EK contributed to the conception and design of the modelling. All authors contributed to the reporting of the work and writing of the manuscript. RS had final responsibility on analysis, reporting and writing. All gave final approval of the version to be published.
- Accepted 5 March 2009
- Published Online First 3 April 2010
Objective To approximate the workload of blood pressure (BP) measurements and lifestyle counselling in primary healthcare when the related guidelines are followed. To evaluate the impact of facilitated guideline implementation with respect to workloads.
Design Modelling study after cross-sectional audit process.
Setting Thirty-one municipal health stations.
Intervention Intrinsic facilitation in implementation of hypertension guideline.
Main outcome measures Number and level (normal <140/85 mmHg, slightly 140–160/85–95 mmHg or markedly elevated >160/95 mmHg) of BP measurements at nurses' appointments, approximation of time allocated for measurements and lifestyle counselling before and 1 year after the intervention.
Results A total of 3119 BP measurements were recorded during the audit week in 2002. BP level measurements were “normal” in 1214 (38.9%), slightly elevated in 1371 (44.0%) and markedly elevated in 534 (17.1%). According to the modelling, 12% of a nurse's workday consisted of BP recordings and counselling. After intervention, the corresponding figures were 2330 measurements (828 (35.5%) normal, 990 (42.5%) slightly and 512 (22.0%) markedly elevated) corresponding to 6.3% of the workday.
Conclusions Through facilitation programmes, it is possible to change working practices according to the related guidelines, agree on the division of tasks and empower patients to engage with their own treatment. These changes can lead to considerable decreases in the workload of health centre personnel with consequent redistribution of personnel resources to patients in true need for services.
Health promotion and disease prevention, such as the follow-up of risk factors for cardiovascular diseases (CVDs) and lifestyle counselling, are important parts of primary healthcare work. CVDs are one of the leading causes of death among the working age population.1
Clinical evidence-based guidelines have been published for the major diseases to assist healthcare professionals in decision-making. Despite a need to base practices on evidence, only a minority of health centres are goal-oriented in the adaptation of guidelines.2 It has been estimated that an average primary care physician in the USA would need 7.4 h of each working day to provide all the preventive services recommended by the US Preventive Services Task Force.3 At the same time, the financial resources available for healthcare are constrained.
Implementing preventive services in primary care is problematic. In a Cochrane systematic review, no solid basis was established for supposing that any particular intervention strategy would work.4 Effective interventions exist, but there is considerable variation in the levels of change. Tailoring interventions to address specific barriers to change is important. Moreover, multifactorial lifestyle interventions are as effective as single ones.5
A facilitator model has proved to be effective in both changing preventive practices and implementing guidelines to organise the prevention of CVDs through general practice.6 However, it is not clear to what extent a greater frequency of risk factor measurements and control are affecting the organisation of primary care.
In this study, we sought to evaluate the time resources needed for blood pressure (BP) measurements and lifestyle counselling when the related guidelines are followed in a city of half a million inhabitants and whether a facilitated local hypertension guideline implementation programme has an effect on the workload.
The Helsinki Health Centre, responsible for 559 000 inhabitants, had 292 general practitioners, with 560 nurses in 31 health stations in 2003. One general practitioner and a multiprofessional team had responsibility for 1800–2500 inhabitants. Per year, approximately 41% of the population used outpatient primary care services. Services for the remaining population were provided by private services, occupational healthcare or secondary care hospitals.
Helsinki Prevention Programme
The Helsinki Prevention Programme, described elsewhere in detail,7 was a 2-year multiprofessional and educational programme. The aim was to improve the prevention of lifestyle diseases and multiprofessional cooperation by creating local guidelines agreeing on the division of tasks. One doctor–nurse pair was recruited from each health station (total n=62) to act as intrinsic facilitators at their own workplaces. The facilitators were coached to identify and introduce early interventions for high-risk (CVDs) patients according to the national hypertension guideline. Educational methods included lectures, workshops, patient cases and role modelling. Quality improvement tools (such as auditing and benchmarking), process methods and skills in leading change were practiced. The facilitators revised local hypertension guideline according to the national Current Care guideline and implemented the guideline in their own health stations (figure 1,2). Self-measurement places were set up to empower patients to engage in their own treatment.
Audit of BP measurements
All nurses performing outpatient consultations and those who voluntarily participated in the auditing were asked to log all BP measurements during one audit week.7 The decision of the nurses to do BP measurement was based on their own assessment—for example, the presence of a chronic disease, major symptom or the wishes of the patients. Patients were grouped in three BP groups, that is, (1) those with “normal” (<140/85 mmHg), (2) slightly elevated (140–160/85–95 mm Hg) and (3) markedly elevated BP levels (>160/95 mmHg). The higher BP figure (systolic or diastolic) determined the group. In addition, the number of patients receiving lifestyle counselling and advised to see a doctor was recorded.
Modelling of time spent on BP measurements and lifestyle guidance
The workload of BP measurements and lifestyle counselling for all nurses during 1 year was calculated based on the 1-week audit sample before and 1 year after the guideline implementation. It was assumed that all nurses conducted a mean number of BP measurements during each single week. The working day (7.5 h) and each week (5 days) of the year were assumed to be similar for each nurse. Four weeks' vacation per nurse was accounted for. The allocations for preventive guidance were based on the Current Care guidelines,8–10 and the US recommendations on preventive activities given in Refs. 3 11 and 12 as follows. The time used for BP measurements was 7 min, for antismoking counselling 3 min, for counselling on physical activity 4 min and for diet counselling 8.20 min. We estimated that patients with normal BP needed only the time required for BP measurements without further personal counselling. General information and health leaflets were always available. Of the patients with slightly or markedly elevated BP, all received both dietary and physical exercise counselling. For both elevated BP groups, 22.8% of the patients were calculated to be smokers (based on a smoking rates audit, unpublished observation) and were thus given antismoking counselling. The time spent on BP measurements and counselling was extrapolated to cover the entire city.
Based on the nurses' (n=172) logs, during 1 week in April 2002, there were 3119 BP measurements altogether. Of the recordings, 534 (17.1%, 95% confidence interval (CI) 15.8 to 18.4) were markedly elevated, 1371 (44.0%, 95% CI 42.2 to 45.7) slightly elevated and 1214 (38.9%, 95% CI 37.2 to 40.6) were normal.
In the follow-up 1 year later (April 2003), nurses (n=250) measured 2330 BPs. Of the recordings, 512 (22.0%, 95% CI 20.2 to 23.7) was markedly elevated, n=990 (42.5%, 95% CI 40.5 to 44.5) slightly elevated and 828 (35.5%, 95% CI 33.6 to 37.5) normal. These results used in modelling are described in detail elsewhere.7
In 2002, the nurses gave lifestyle counselling to 1095 (35%) patients and 223 (7%) patients were advised to see a doctor. The corresponding figures after the 1-year follow-up in 2003 were as follows: 992 (42.6%) and 236 (10%).
Workload modelling based on performance according to national guidelines
The mean number of BP measurements for one nurse was 18 recordings per week and the approximation of BP recordings during 1 year by all nurses in the city of Helsinki corresponded to 487 000 measurements. Furthermore, the approximated annual rate for visits because of BP recordings for nurses was 2100 visits per 1000 registered patients. Corresponding figures after the follow-up were 250 500 measurements per year and 1100 visits per 1000 registered patients. Approximations of the time spent on BP measurements and lifestyle counselling are shown in tables 1 and 2. According to these approximations, the working hours (37.5 h/week) of 66 nurses (normal BP 12, slightly elevated BP 39 and markedly elevated BP 15) were used solely for measuring BP and providing lifestyle counselling in Helsinki in 2002. The corresponding figures in 2003 would have been a total of 36 nurses for the different BP levels (normal BP 6, slightly elevated 20 and markedly elevated 10, respectively).
Properly performed BP measurements and lifestyle counselling of these patients can account for 12% of the working hours in a city of half a million inhabitants. We observed that by changing working practices according to the national hypertension guideline and by the agreement on task division (local guideline), it is possible to gain time benefits for redistribution of resources and target treatment to the relevant patients.
The effectiveness of guideline implementation is difficult to study because often implementation processes are dependent on managerial processes and constitute quality projects rather than pure healthcare research. The strength of our study lies in its revealing the unselected patient population and everyday work in primary care. During the follow-up, the patient coverage of each geographical practice (41.2%) and the number of employees remained unchanged. The turnover of employees was maximum 12%. The weakness of our study lies in the estimation being performed using a 1-week sample. Because the samples were large, we assume that overestimation of the workload is unlikely. On the contrary, an underestimation may have occurred because BP measurements and counselling conducted in maternity clinics and homecare were not included in our approximations. In addition to avoiding overestimations, we used smoking rates based on a local audit, lower than in studies on national frequencies of smoking13; and in the modelling the smoking cessation, counselling was directed only at patients with elevated BP.
Many chronic diseases, including CVDs, are caused or influenced by a variety of risk factors that should be targets of guidance by general practitioners. A number of studies have shown that, like other cardiovascular risk factors, hypertension is common and undertreated.13–17 Physicians alone cannot deliver all the services recommended by even “conservative” sets of preventive “task forces” without collaboration of other professionals.3 Well-functioning healthcare teams demonstrate better patient and organisational outcomes.18–20 A key to such an improvement lies in the role of nurses in activating patients in self-care.21 22 Guidelines do not usually include directions on who is the caregiver. A simple quality improvement programme can improve teamwork, and the practice of recording risk factors in CVD result in the earlier detection of patients with high risk of the disease.23 It would be possible to improve clinical practices and support proactive practice teams further through local guidelines accompanied by an agreement on task division.
However, we still do not know how time consuming it is to follow and counsel these patients. In the present study, the time approximations used for monitoring BP and counselling these patients are based on accepted national and international estimates.5 8–12 Based on these approximations of time loads used for measurements and preventive discussions in primary healthcare sector in a middle-sized capital, we suggest a workload of 10–15% of the daily working time of each nurse. The number of measurements diminished by one-third during the follow-up. The decrease in itself does not mean better care. However, the focus of measurements shifted to those with poor treatment control. This was the expected pattern of change and improvement. According to the guideline, it is important to take into account the total risk and target the treatment activities to those in poor treatment control. The potential confounder of a shift of BP measurements between physicians and nurses cannot be ruled out. Still, the intervention programme particularly aimed at avoiding overlapping duplication of routine work. This should result in a more responsible role of the nurses in BP measuring and lifestyle counselling. Therefore, the shift was from physicians to nurses and from nurses to patients' self-measuring. Three components of the Helsinki Prevention Programme were similar to the Chronic Care Model.24 Self-treatment places were set up to support self-care, and local guidelines (task division) did result in the redesign of the delivery system while providing support in clinical decision-making and management. Attention was refocused from motivated patients with good treatment control to those at high risk of CVD.
Observed changes can be translated into rough estimates of the financial benefits. An appointment with a nurse costs €31 (2003). Based on our study, there would have been 237 000 fewer BP appointments, corresponding to redistribution in expenses of €7.3 million per year.
We suggest that primary healthcare workers, that is, general practitioners and nurses, give serious consideration to sharing the practice of BP measurements and lifestyle counselling for patients with CVD risk factors. Through this new division of tasks and empowerment, it would be possible to better manage working hours. Furthermore, we suggest that experts presenting national guidelines take the workload and financial consequences into consideration.
We thank the facilitators for organising the audit process.
Funding Grant from the Helsinki city health department, the Wilhelm and Else Stockmann Foundation and the Finnish Medical Society Duodecim.
Competing interests None.
Ethics approval This study was conducted with the approval of the ethical committee of epidemiology and public health, hospital district of Helsinki and Uusimaa.
Provenance and peer review Not commissioned; externally peer reviewed.