Intended for healthcare professionals

Primary Care

Qualitative study of pilot payment aimed at increasing general practitioners' antismoking advice to smokers

BMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7310.432 (Published 25 August 2001) Cite this as: BMJ 2001;323:432
  1. Tim Coleman, senior lecturera (tjc3{at}le.ac.uk),
  2. Alison T Wynn, research associatea,
  3. Keith Stevenson, lecturera,
  4. Francine Cheater, professor of public health nursingb
  1. a Department of General Practice and Primary Health Care, Leicester Warwick Medical School, Leicester General Hospital, Leicester LE5 4PW,
  2. b School of Health Care Studies, University of Leeds, Leeds LS2 9LN
  1. Correspondence to: T Coleman
  • Accepted 21 May 2001

Abstract

Objectives: To elicit general practitioners' and practice nurses' accounts of changes in their clinical practice or practice organisation made to claim a pilot health promotion payment. To describe attitudes towards the piloted and previous health promotion payments.

Design: Qualitative, semistructured interview study.

Setting: 13 general practices in Leicester.

Participants: 18 general practitioners and 13 practice nurses.

Results: Health professionals did not report substantially changing their clinical practice to claim the new payments and made only minimal changes in practice organisation. The new health promotion payment did not overcome general practitioners' resistance towards raising the issue of smoking when they felt that doing so could cause confrontation with patients. General practitioners who made the largest number of claims altered the way in which they recorded patients' smoking status rather than raising the topic of smoking more frequently with patients. Participants had strong negative views on the new payment, feeling it would also be viewed negatively by patients. They were, however, more positive about health promotion payments that rewarded “extra” effort — for example, setting up practice based smoking cessation clinics.

Conclusions: General practitioners and practice nurses were negative about a new health promotion payment, despite agreeing to pilot it. Health promotion payments do not automatically generate effective health promotion activity, and policymakers should consider careful piloting and evaluation of future changes in health promotion payments.

What is already known on this topic

Health promotion payments have been made to UK general practitioners since 1990, but their effectiveness is unknown

What this study adds

Primary care staff held strong negative views about the pilot payments to promote smoking cessation and previous health promotion payments

The highest claiming practitioners altered their methods of recording smoking status rather than increasing the frequency with which they advised patients against smoking

Future changes in health promotion payments should be carefully piloted

Introduction

Paying general practitioners for various activities has been shown to affect their behaviour.16 Since 1990, the UK government has used payments for specific general practice clinical activities with varying success. For example, target payments have probably helped to improve rates of immunisation2 and cervical cytology.7 However, health promotion payments, which were introduced in the United Kingdom in 1990,8 have been less successful. 9 10 They have usually been introduced without piloting or any evaluation to determine whether they change clinical behaviour. This is unfortunate because general practitioners perceive many barriers to the effective practice of health promotion within their routine consultations that payments may not easily overcome.1114

We piloted a new health promotion payment aimed at increasing general practitioners' antismoking advice.15 In this paper, we investigate the changes in clinical practice and practice organisation reported by general practitioners and practice nurses in order to claim the new payments. We also investigated their attitudes towards the new and previous health promotion payments.

Participants and methods

Thirteen general practices (out of 28 approached) in one sector of Leicester agreed to pilot a new health promotion payment. Practices were paid £15 for each patient identified who had smoked during the past 12 months but was not doing so now and had stopped continuously for at least three months. Claim forms could be completed by any member of practice staff but had to be signed by a general practitioner. Practices were studied before and after introduction of the payment.15

Thirty five general practitioners and 18 practice nurses worked in the participating practices. We aimed to interview at least one general practitioner and practice nurse from each participating practice, sampling interviewees with a wide range of attitudes towards the new health promotion payments.16 We interviewed one practice nurse from each practice, choosing the person whose work was most likely to include engaging smokers.

As there were more general practitioners, we were purposive in selecting them for interview.16 Before payments were introduced we selected general practitioners of different ages and sex and also some who had attended recruitment meetings and some who had not.15 We aimed to sample general practitioners with different views on stopping smoking and financial incentives. After payments were introduced, we selected general practitioners on the basis of the number of claims submitted, aiming to include a full range. We also sought “atypical” general practitioners who were making many claims but who were working in practices where others made relatively few. The final categorisation of practices and individuals by numbers of claims was done at the end of the intervention period.

Development of interview guides

We developed semistructured guides for the interviews with general practitioners and practice nurses.16 The general practitioner guide was based on a literature search, unpublished data from another study,12 and pilot interviews. The practice nurse guide was developed after literature review, discussion with a practice nurse facilitator for the health authority, and a focus group meeting of practice nurses working in non-participating practices.

Interviews covered the following subjects: usual approach to discussing smoking, expected or actual organisational changes made by practices to claim payments, expected or actual changes in clinical practice made as a consequence of claiming payments, and attitudes to the new and previous health promotion payments. We conducted similar numbers of interviews before and after introduction of the payment to document expected changes as well as actual change. We hoped that interviewing at different points in the piloting process would enable us to understand decision making within practices. We interviewed practice nurses because they often have a prominent role in health promotion within practices and could provide a different perspective on how practices responded to pilot payments. Interviews lasted 35 to 75 minutes and were audiotaped and transcribed verbatim.

Analysis of data

The box summarises the process of data analysis. The themes and categories on which our analysis is based were derived from the data, rather than being imposed by the researchers.17 The definitions of the emergent themes and categories were checked against the data and subsequently refined in an iterative process.18 Details of working definitions, revisions, and final definitions of themes and categories are available.19 During the final coding, this iterative process continued, and interview text was compared with written descriptors. In a few instances, the data did not fit into existing definitions; we then refined and agreed the definitions. We aimed to describe the main themes emerging from interviews and not to develop a typology for categorising individual general practitioners or practice nurses.17 After coding all interviews, we collated interview text from themes and categories relating to the research question to identify important issues. Finally, FC and KS read a subsample of five randomly selected transcripts and confirmed that these contained data that supported the principal findings.

Process of analysing data

TC and AW independently read the first 16 interview transcripts to identify themes (principal issues relevant to new health promotion payments)

TC codes transcripts for themes. Data relating to each theme assembled. TC and AW independently read this and identify categories (variations in thinking within themes)

TC and AW agree working definitions for emerging categories and themes

TC codes the 16 transcripts using these definitions

TC and AW discuss and refine definitions of themes and categories

TC begins recoding all available and subsequent transcripts using agreed definitions. During this process some definitions are altered after discussion between researchers

KS and FC read random sample of manuscripts to assess whether coding is adequate

Results

We present qualitative results in the context of the number of claims made by participating practices. Two practices were high claiming (mean number of claims per doctor ⩾11), five were medium claiming (3 to 10) and six were low claiming (⩽2). Thirty one interviews were conducted in total, of which five were with staff from high claiming practices (three doctors, two nurses), 14 with staff from medium claiming practices (nine doctors, five nurses), and 12 with staff from low claiming practices (six doctors, six nurses). We interviewed nine doctors and seven nurses in the control period and nine doctors and six nurses in the intervention period. All 13 practice nurses and five of the 18 general practitioners were women. TC conducted 21 of the interviews (four with practice nurses) and AW conducted the rest. Of the general practitioners, five had worked in their practices for more than 10 years, 10 for 5-10 years, and two for less than five years. For practice nurses, the figures were one, eight, and two years respectively (data missing for two).

Expected changes and barriers to change

Some participants thought that the introduction of the payment would prompt them to broach the topic of smoking more frequently, but many indicated that potential earnings from the new payments were insufficient to produce major changes in either clinical practice or practice organisation.

I think we kind of, were quite keen until we sort of worked out what the numbers might be and we thought, “Hang on, this is going to be hardly anybody [[to claim about]] so probably we are going to end up no better off.” (doctor in medium claiming practice)

Despite expecting to discuss smoking more frequently, many doctors indicated that they would be unlikely to change the manner in which they raised the issue of smoking. Practice nurses did not anticipate changing their clinical behaviour. One doctor suggested she would still not discuss smoking with the patient unless it was “relevant to the patient's reason for consulting.” Others indicated that they would continue to discuss smoking primarily in the context of smoking related problems or when smokers seemed motivated to try stopping. Some participants felt it more important to avoid confrontation with smokers than to try to achieve financial gain by discussing smoking with them:

It seems they resent you asking about something that's totally unrelated to the reason they've come to see you. (doctor in medium claiming practice)

Participants also realised that claims could easily be made without doctors and patients discussing smoking, and some were considering making claims solely based on audits of medical records.

Only one practice reported discussing and implementing systematic changes in its clinical practice in order to get payments. This practice, which was high claiming, introduced systematic questioning of patients about their smoking habit and systematic recording of this information on its computer system. Smokers who had stopped, or expressed an intention to stop smoking, were subsequently identified from computer records and contacted by practice staff to determine whether claims could be made. A single handed general practitioner from the other high claiming practice indicated (before claims were introduced) that he was considering starting to record smoking status systematically to make claims.

The two atypical high claiming general practitioners reported a systematic approach towards discussing smoking before payments were introduced. One reported that he had previously documented all antismoking advice given to patients, but when the payment was introduced he began highlighting smoking status to ensure that he remembered to raise the issue and claim where possible. The other routinely prescribed nicotine replacement therapy to patients, both before and during piloting of the new payment. He regularly reviewed patients who took nicotine replacement therapy and, where appropriate, made a health promotion claim. High claiming and atypical general practitioners indicated that making claims provoked little or no change in the way they raised the issue of smoking. They also reported preferring to raise the topic of smoking “as appropriate” and with smokers who were “motivated to stop.” There was little evidence that the payment generated changes in the clinical management of smoking cessation in the practices of these general practitioners (box).

Reported changes in high claiming practices and atypical high claiming general practitioners

“My practice manager was given a list of patients who fulfilled the criteria [[for claiming]] but as yet we've had great difficulty either chasing them up or finding out whether they've still stopped smoking.”

“We discussed setting up the separate clinics, inviting people to come along, but as time went on it became more apparent that it just would not be viable, with the level of payment, to put an awful lot of work into it. We are all agreed on that now. What will happen in the run-up to September [[when the GP thought the new payments became available]] is we are identifying as many people as possible who are smokers and over the following months from September we will find out who stopped and at what point and obviously press on with encouraging people to stop during the course of the working day.”

“I think in terms of patient care, if I'm going to carry on giving the advice then the benefits to them are going to be the same whether I document it in capitals and follow it through, I'm not concerned. So the only reason why I would carry on documenting is if there's a financial remuneration for the practice, because it's a paper exercise for the benefit of auditors and the health authority, and in return I would expect to be paid for that.”

Attitudes towards pilot payments

Almost all participants were negative about the payment. Very few doctors and no practice nurses felt that the payment could improve the management of smoking cessation. The doctors with positive views tended to modify their responses with negative qualifications. Many participants thought the payments were unfair because they believed it was more difficult to change the smoking behaviour of patients from socioeconomically disadvantaged backgrounds. Most doctors and nurses felt the payment did not reward the effort that primary healthcare teams put into health promotion.

If you were being paid because people have stopped smoking, then to some extent whether they stop smoking or not is slightly out of your control. You do your best, but whether they actually do stop smoking is really, at the end of the day, down to them.

I don't mind the money being taken away if it is going to be given to the practices where they are making an effort, but I do just think that the whole claims thing could just be a disaster where organised practices are making money out of nothing and making no effort at all on the smoking front.

Some questioned the payments on ethical grounds, believing that giving lifestyle advice should be part of their contractual obligations:

I think [[giving antismoking advice to patients]] is a contractual obligation and a moral issue. You are a health professional. I look at myself as a health professional, not only diagnosing disease. (doctor in low claiming practice)

Concern was expressed that the payment could promote inappropriate clinical activity. If the issue of smoking was raised too frequently, it would lessen the effectiveness of intervention. Some felt it would be particularly difficult for health authorities to validate claims, making it impossible to audit new payments.

When asked, both general practitioners and practice nurses agreed that it would be fairer to receive payments for providing additional services to smokers that involved time and resources being dedicated to them. This could involve supporting individual smokers who were prepared to try to quit or running practice based smoking cessation clinics.

Perceptions of patients' reactions

Most doctors and nurses thought that patients would have negative attitudes if payments were introduced. They predicted vigorous reactions, using words like outrage and resentment. Participants thought that patients would be unhappy about doctors being financially rewarded for identifying whether patients had stopped smoking and that this could harm doctor patient relationships. No one thought that patients would have completely positive reactions to the introduction of the payment.

Views on previous health promotion payments

The box summarises health promotion payment schemes introduced since 1990. Virtually all participants were negative about payments for health promotion clinics. Only two had positive comments: one practice nurse liked the systematic approach to health promotion that clinics encouraged and one general practitioner was positive about the income that could be made from health promotion clinics:

Health promotion payment schemes introduced since 1990

1990-3: Fee for service payments made for delivery of approved health promotion activity within clinics based in general practice and attended by a set number of patients5

1993-6: Target payments introduced for recording cardiovascular risk factors in a given proportion of the practice population. Also known as the banding system6

1993-6: Fee for service payments made for clinics dedicated to treatment of diabetes and asthma

1996 onwards: Banding scheme scrapped but payments for asthma and diabetic clinics retained. Practices now reimbursed for locally approved health promotion activities. No national health promotion payment scheme exists

Yes, so we made a lot of money very quickly out of that and silly clinics saying to your hypertensives, “all come on a Thursday” and you'd be ten booked pressure checks, and booking a claim for it. It was a total con to be honest, but it bought me a new car.

Health promotion clinics were considered difficult to organise and inflexible for patients to attend. Clinics were perceived as alien to the usual culture of general practice, which involves patients attending surgery when they have health problems. Some participants thought that clinics attracted the “worried well” and not those who were most likely to benefit. Advice given in clinics was perceived to be less effective than that given during routine consultations:

If people come with a problem I think they take more notice of your advice than if they are coming to a [[health]] promotion clinic.

One of the strongest objections to the clinic based health promotion payment system was that it encouraged inappropriate clinical activity and wasted NHS resources. Participants reported working in or knowing about practices that asked patients to attend health promotion clinics more frequently than was necessary, merely to generate clinic throughput and claim fees.

Few practice nurses expressed opinions on payments for recording cardiovascular risk factors because most had not collected data for this. Some general practitioners, however, thought that this system was better than clinic based health promotion payments. These more positive views were largely based on the ease with which data collection could be delegated to others.

You could do it largely opportunistically. You could delegate a lot more of it to non-medical or at least, you know, your nursing staff, administrators, because it is a lot of work that we don't really need to be doing. It can be done opportunistically. It could be monitored sensibly, and it would be an ongoing review of payment. If you were found falling behind you would have to, you know, instigate administrative or medical matters to pull yourself up.

Another positive view of the cardiovascular risk scheme was that it encouraged recording of information on health promotion in medical records.

Nevertheless, the scheme produced far more negative comments than positive responses. Doctors often reported that the data required by health authorities were time consuming and difficult to collate. Retrospective audits of medical records were often used to compile data, suggesting that clinical behaviour was not changed. Doctors had received no feedback from their health authority about information collected, resulting in widespread cynicism about the value of the exercise. Several people had concerns about the quality of data collected, with one indicating that the data were of such little value that no effort was justified to ensure its accuracy.

Well it took about, I don't know how many hours the first year. So obviously when nobody bothers to check up on it you tend to make it up a bit.

Discussion

The new health payment did not changes practice nurses or general practitioners' behaviour towards advising patients on stopping smoking.15 Our findings suggest that primary care staff remained resistant to raising the topic of smoking more frequently because they thought this could engender confrontation with patients. Payment did not overcome this previously documented fear.12 General practitioners also did not feel the payment was sufficient to reflect the work needed to identify patients who had recently stopped smoking. Consequently, practices made few organisational changes to claim the payment, and clinicians did not change their clinical practice greatly. The general practitioners who made the most claims integrated claiming into existing clinical behaviour rather than altering behaviour to claim payments. These general practitioners did not report raising smoking more frequently but admitted altering recording practices to document patients' smoking behaviour.

Many participants were strongly against wider introduction of payments as they anticipated strong negative reactions from patients. This suggests that increasing the payment would not make it more acceptable. General practitioners and practice nurses were more positive towards health promotion payments that rewarded extra effort that was not part of their “core” work. Additional payments made to practices for supporting smokers who are motivated to stop smoking or perhaps for setting up practice based smoking cessation clinics might be more acceptable. An expert group has recommended the UK government to reimburse practices prepared to provide extra support to smokers who are motivated to stop,20 but the demand for community support for smoking cessation is unlikely to sustain a cessation clinic in every general practice.21

Validity

Recent health promotion payment schemes were unpopular and may have generated inappropriate activities rather than effective health promotion. We cannot be certain about the actual activities that previous health promotion schemes encouraged, but the message that the cardiovascular risk scheme promoted administrative changes and changes in data collation are consistent with other findings. 9 22

We have probably obtained a fairly complete picture of how general practitioners and practice nurses viewed the pilot payments. We selected participants who would be expected to have varied attitudes and allowed them to express their views freely. Nevertheless, we could not validate explanations given by interviewees. In addition, health professionals in practices that declined to take part may hold different views.

Need for evaluation

Our study shows the value of evaluating new financial incentives to general practitioners using qualitative research methods. When health promotion payments are made available, general practices seem to follow the path of least resistance to claim them. This often involves simple administrative changes rather than changes in clinical behaviour. New payments for health promotion should be carefully piloted and evaluated to determine whether they alter clinical practice.

Acknowledgments

We thank Margaret Whatley and Pauline Green for secretarial help, Toby Gosden for comments on an earlier version of this report, the 13 general practices that participated in the study, and Orest Mulka, who proposed the new health promotion payment.

Contributors: TC obtained funding for the study. All authors contributed to interview design and data analysis. TC and ATW conducted interviews, TC wrote the paper, and all authors commented on drafts. TC and ATW are guarantors for the paper.

Footnotes

  • Funding Trent NHS Executive and Medisearch.

  • Competing interests None declared.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.