Since 1990, the UK government has tried to influence health promotion activity by general practitioners through payment schemes.1 These have never been rigorously evaluated.2 We examined the feasibility and effectiveness of a payment scheme that aimed to increase general practitioners' antismoking advice in an uncontrolled before and after study.
Participants, methods, and results
The health promotion payment was piloted in a deprived area of Leicester. The recruitment of practices is described elsewhere.3 Thirty five general practitioners (out of 62 approached) from 13 general practices (out of 28 approached) were recruited, and 31 participated in the study.
Before data collection began, we invited all members of primary healthcare teams to attend training in methods of stopping smoking. We then observed normal clinical behaviour over nine months (the control period). In the following nine months (the intervention period), practices could claim £15 from the health authority for identifying each patient who had smoked during the past year but was currently not smoking and had not done so for at least three months. We estimated that individual general practitioners could claim between £285 and £1125 annually.
We needed to recruit 904 smokers to measure a 10% absolute change in the proportion of smokers receiving antismoking advice from their general practitioner, with 80% power at a 5% significance level. A research assistant asked all patients (parents or guardians of those <16 years) attending a random selection of general practitioners' surgeries to complete a questionnaire before the consultation. This sought sociodemographic details, identified regular smokers (those smoking on, at least, most days), and asked about smoking behaviour and smoking related problems. Smokers were asked to complete a second questionnaire after the consultation, asking whether they had been given antismoking advice. Patients who could not complete the questionnaires were excluded. We compared the researcher's records with those of receptionists to estimate the number of missed patients.
We compared the proportions of regular smokers who recalled discussion of smoking with their general practitioners before and after introduction of payments using the Mann-Whitney U test and allowing for clustering of data.
The table shows that patients in the intervention group were older and more motivated to stop smoking than those in the control group but that the distribution of smoking related problems was similar in both groups. We found no significant difference in the proportion of smokers recalling general practitioners' antismoking advice before and after introduction of the payment.
The numbers of smokers seen by each general practitioner (cluster size) varied greatly, and the proportions of smokers recalling antismoking advice were not normally distributed (intercluster correlation coefficient for recall of antismoking advice=0.052). Fourteen doctors made no claims, 15 made one to nine claims, and four made over 10.
Comment
Paying general practitioners to identify smokers who had stopped smoking for three months or more did not make them give antismoking advice more frequently. The reasons behind the failure of the payments to change behaviour are explored elsewhere.3
Our findings could have been influenced by external factors, and offering smoking cessation training before the study started may have increased the amount of advice given during the control period.4 This would make it difficult to detect a small effect of the payment. Differences between the control and intervention groups at baseline are unlikely to account for our findings.5 We have no evidence to argue for a cluster randomised control trial of this payment scheme.
We have shown that it is feasible to investigate the introduction of a general practice health promotion payment in a prospective, experimental study. Future payment schemes can and should be evaluated using experimental methods.
Table.
Comparison of patients in control and intervention periods. Values are numbers (percentages) of patients unless stated otherwise
| Characteristic (No answering question) | Control period | Intervention period | P value for difference |
|---|---|---|---|
| Completed first questionnaire | 1601/1878* (85.3) | 1354/1647† (82.2) | — |
| Women | 1064/1601 (66.5) | 860/1354 (65.5) | 0.09 |
| Mean age (years) (n=2928) | 44.0 | 45.3 | <0.05‡ |
| Regular smokers (n=1026) | 527/1601 (32.9) | 499/1334 (36.9) | 0.03§ |
| Had tried stopping in past year (n=1013) | 205/518 (39.6) | 218/495 (44.0) | 0.15§ |
| Intended to quit in next month (n=1007) | 97/515 (18.8) | 137/492 (27.8) | 0.001§ |
| Confident of being able to stop (n=1010) | 138/517 (26.7) | 156/493 (31.6) | 0.08§ |
| Want to stop (n=1004) | 268/513 (52.2) | 297/491 (60.5) | 0.008§ |
| Thinking about or trying to stop (n=1004) | 209/513 (40.7) | 218/491 (44.4) | 0.24§ |
| Seeing general practitioner about themselves and perceive they have a smoking related problem (n=806) | 75/396 (18.9) | 80/410 (19.5) | 0.84§ |
| Completed second questionnaire | 466/527 (88) | 461/499 (92) | 0.03§ |
| Recalled antismoking advice | 100 (21) | 87 (19) | 0.62¶ |
6 patients excluded, 226 missed, and 45 refused to participate.
10 patients excluded, 231 missed, and 52 refused to participate.
By t test.
By χ2 test.
P value for median difference in percentages by Mann-Whitney U test.
Acknowledgments
We thank Margaret Whatley for secretarial help and Keith Stevenson and Toby Gosden for commenting on earlier versions of this paper. We also thank Tina Booth and Jane Roberts, who provided health promotion training, and Leicestershire Health, especially Iain Harkess and Clifford Hughes, who helped complete the main study. Finally, we wish to thank the 13 practices that participated in the study and Orest Mulka, who had the original idea for the health promotion payment.
See p 432
Footnotes
Funding: Trent NHS Executive and Medisearch.
Competing interests: None declared.
References
- 1.Department of Health. General practice in the national health service: a new contract. London: HMSO; 1989. [Google Scholar]
- 2.Giuffrida A, Gosden T, Forland F, Sergison M, Leese B, Pedersen L, et al. Target payments in primary care: effects on practice and health outcomes. In: Bero L, Grilli R, Grimshaw J, Oxman A, editors. Collaboration on effective professional practice module of the Cochrane Database of Systematic Reviews. Cochrane Library. Oxford: Update Software; 2000. . CD000531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Coleman T, Wynn AT, Stevenson K, Cheater F. Qualitative study of pilot payment aimed at increasing general practitioners' antismoking advice to smokers. BMJ. 2001;323:432–435. doi: 10.1136/bmj.323.7310.432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lancaster T, Silagy C, Fowler G. Training health professionals in smoking cessation. Cochrane Database Syst Rev 2000;(2):CD000214. [DOI] [PubMed]
- 5.Coleman T, Wilson A. Factors associated with provision of anti-smoking advice by general practitioners. Br J Gen Pract. 1999;49:557–558. [PMC free article] [PubMed] [Google Scholar]