Your browser has 'Cookies' disabled, alert boxes will continue to appear without this feature.

Putting prevention into practice (Green Book)


There is a growing literature on what implementation strategies are effective. Nevertheless there are a number of continuing paradoxes that relate to the clinician’s understanding about implementation.

Strategies most preferred by clinicians often have the least impact. Most traditional continuing professional development evenings that include a visiting specialist speaker over a dinner meeting have minimal impact on clinician performance.177-179 Nevertheless, they are popular as they rarely require additional work or effort on the part of the clinician. On the other hand, organisational strategies usually have a large and consistent impact.

The corollary of this is that interventions offered to clinicians where their self reported performance is poor, often have a bigger impact than interventions aimed at improving their performance when it is well above average.

According to Grimshaw,180 passive dissemination is generally ineffective. More active approaches such as reminders and educational outreach are more likely to be effective but are also more costly. Interventions based on an assessment of potential barriers to change are most likely to be effective. And multifaceted interventions targeting different barriers are likely to be more effective than single interventions.181,182

Consequently, it is important to undertake a systematic planning process to identify barriers and enhancers of preventive activity in your practice before launching into a program of activity (Table 6).




Organisational strategies (eg. clarification of roles, delegation of tasks, practice policy/standing orders, protocols, incentives)

Highly effective

Contributes to implementation of preventive interventions and helps sustain them
Impact varies with area, capacity and acceptability

Reminders for the GP

Very effective

Computerised reminders have a similar impact to manual reminders. Needs to be targeted

Reminders for patients

Very effective

Needs to be targeted

Other interventions and reminders for patients

Very effective

For example, telephone, patient education, support strategies

Practice nurse interventions


Provides a clear outline of the role of the PN and gives adequate training and support

Practice co-ordinator


May be someone within the practice or external

Health summary sheet


Practice accreditation standards require a minimum number to be completed

Case note audit


Impacts particularly on prescribing and test ordering

Continuous quality improvement


Needs active GP involvement and feedback, and a supportive practice infrastructure



More effective for conditions involving a team of health professionals and where large numbers of patients need to be seen


Effective in some situations

Needs to be pre-negotiated and tailored. Peer comparison is useful if confidential

Practice registers

Effective in some situations

Require a computer to be most effective

Local opinion leaders

Effective in some situations

Assist in spreading information and examples


Not effective


Traditional CME evenings

Not effective


Table 6. The effectiveness of implementation strategies in improving prevention

Grimshaw and colleagues, in a recent systematic review of interventions aimed at changing clinician behaviour and/or performance, found there was no significant effect in size (improvement in performance) with increasing the number of interventions to facilitate implementation (Figure 8). More is not necessarily better. More importantly, it is the strategic combination of implementation strategies that is the key to improving performance and not just the use of multiple approaches.

Figure 8. Effect sizes 
of multifaceted interventions by number of interventions

Figure 8. Effect sizes of multifaceted interventions by number of interventions
Reprinted with permission: Grimshaw J, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Healt Technol Assess 2004;8:1–72

We often assume that the effort to implement an activity is the same regardless of the level of initial performance. As seen earlier with the brief advice for smoking cessation, the impact of GP advice falls off sharply after 3–5 minutes, even though it continues to improve. Specifically, the return on effort is not linear. This is further demonstrated in Figure 9. It is useful to think about three performance ranges:

1. No or very low level of performance
Considerable effort is usually needed to improve low levels of performance or overcome inertia. Similarly low levels of performance should prompt the GP to think about the various constraints impeding performance. For example, the GP’s time is often the constraint in achieving very high levels of prevention coverage. Given the heavy GP workload, adding additional prevention tasks are not attractive, unless than can be performed by someone else.

A similar approach of identifying constraints can be adopted in the assessment of the patient’s motivation or confidence. Low scores on both should prompt the GP to ask what would need to happen to improve this score from, say 2, to 8 or 9? When performance is low, there is likely little infrastructure or an absence of critical mass to support it. Think carefully about what is constraining performance.

2. Mid-range performance (~20–70%)
Improvement in this range is easier up to a point but may not be linear, as described above.

3. High level of performance (70–100%)
Achieving very high levels of performance is influenced by two further principles:

  • the law of diminishing returns: even higher efforts are required to increase the effect by a given percentage, resulting in a reduced efficiency
  • the Pareto principle: 20% of your patients will require 80% of your effort.

In practical terms, this means that if your current performance is very high, say around 80–85%, then a lot of effort will be required to reach 95–100%. Similarly, if your performance is extremely low, then there are likely to be a large number of constraints holding back your performance. These principles provide further support for being strategic in your approach to implementation. Practices vary significantly in terms of resources, infrastructure and patient population profiles and this has a significant influence on preventive interventions.

Consider the following:

  • Is the prevention activity important? (burden of illness)?
  • Am I likely to be effective? (role, impact)?
  • What combination of implementation strategies is likely to be effective?
  • Can I identify the various barriers and constraints to better performance?
  • Can I make the impact and outcome visible?
  • What will assist getting a quick return?
  • Is it desirable?
  • Is it do-able?
  • Can we make it a routine part of the practice?
  • What is the capacity of the practice to provide the intervention? How will implementation strategies work in my practice?
Figure 9. 
Effect-performance paradox

Figure 9. Effect-performance paradox

Return on effort – pneumoccocal polysaccharide vaccine

Less than a third of the elderly receive the pneumoccocal polysaccharide vaccine (PPV). What approach will result in the best return on effort?:

  • Offer the PPV to all elderly patients attending the practice for flu injection. Only a small improvement would be seen if GPs offered the PPV to all patients who came for a flu injection. Many of this group have already had the PPV in the past 5 years
  • Provide PPV free to the elderly and make it available to GPs in their surgeries. This had a noticeable impact when the same strategy was adopted for the flu injection several years ago. It removes any financial barrier and facilitates opportunistic provision of PPV. The majority (>90%) visit a GP at least once in the pre-influenza period so in theory, nearly all could be offered the PPV. The NIPS survey showed that coverage could improve by about 10% using this strategy
  • Flag the case notes of all the elderly patients and ensure that the PPV is recommended to elderly patients when they attend the surgery. The practice staff could inform the PN that an elderly patient attends the surgery and needs the PPV. Alternatively, they could remind the GP. The PPV is provided to patients and receipt of the PPV is recorded on the electronic record so the prompt disappears.

As the influenza season approaches, there will be fewer prompts on the screen or uncompleted vaccine status on the case notes to remind the practice staff or GP. In May, practice staff could generate a list of all elderly patients and whether they had had the PPV. A decision could then be made whether to phone this group and discuss the PPV or wait until they attend the practice. PPV coverage of around 80% could be achieved using this approach. The significant improvement in impact is due to the strategic approach that is used to tackle some of the key constraints to improving PPV coverage. These include: difficulty identifying the target group, systematic approach to increasing coverage, provision of a reminder to the GP and practice staff, and a recognition that a GP recommendation to get the PPV overcomes most of the concerns and misperceptions about the PPV.183

John Litt, Flinders University, South Australia

Key messages

  • Focus on what GPs and the practice are interested, competent, prepared and able to do
  • Build teamwork within your practice and enhance collaboration with other community services. This will greatly enhance your prevention activities and benefit to patients
  • Referring to relevant community services and programs reduces work pressure on the GP
  • Seek like-minded partner organisations to work on preventive strategies
  • Work with and through your local division of general practice