Journal of Public Health Medicine
Vol. 23, No. 4, pp. 329–334 Printed in Great Britain
Obstacles to influenza immunization in primary care Tim Doran and Rosemary McCann
Abstract Background General practices undertake annual immunization campaigns to protect susceptible patients against influenza. Many practices, however, do not adopt effective approaches and there is great variation in the immunization rates achieved. This study aimed to assess the attitudes of primary care staff to the annual immunization programme, the obstacles they face, and possible reasons for the wide variation in immunization rates. Method A semi-structured questionnaire survey of general practice groups in Salford & Trafford during winter 1997–1998 was carried out, a total of 104 practices. Results Respondents perceived influenza vaccine to be effective (93.2 per cent), well received by patients (91.7 per cent) and without significant side-effects (83.6 per cent). The annual immunization programme was seen as being necessary (91.8 per cent), cost-effective (76.7 per cent), reducing hospital admissions (82.2 per cent), but very time consuming (64.4 per cent). Practices were more likely to target patients specified in the Chief Medical Officer’s guidelines; however, most (98.6 per cent) targeted the over-75s before their inclusion in the guidelines, and many (61.6 per cent) targeted the over-65s. Practices did not always use the most effective methods of contacting patients, primarily relying on posters (97.3 per cent), opportunistic contacts (95.9 per cent) and reminders on prescriptions (83.6 per cent), rather than letters (39.7 per cent) and telephone calls (11.0 per cent). Practices identified several common obstacles to immunization, relating to the cost and administrative burden of the annual immunization programme, difficulty identifying high-risk patients, and public beliefs about influenza and influenza vaccine. Conclusions Although generally positive towards influenza immunization, practice staff differ in their ability to cope with, and the appropriateness of their response to, the pressures of the annual immunization programme. Additional support and co-operation from the Department of Health, Health Authorities and pharmaceutical companies could remove some of the obstacles to immunization of high-risk patients. Keywords: influenza, immunization, general practice
and administering the vaccine lies primarily with general practitioners (GPs). Despite advice from the Department of Health on who to immunize (Table 1) and how to run a successful immunization campaign,1 many practices do not adopt effective approaches and there is great variation in the immunization rates achieved. Research has demonstrated that this variation is in part related to GPs’ attitudes towards influenza vaccine and the methods employed by practices to target, identify, contact and immunize patients.2–6 This study aimed to assess the attitudes of general practice staff to the annual immunization procedure, their methods, and the perceived obstacles to appropriate immunization. The study pre-dates the inclusion of the over-75s (in 1998) and the over-65s (this year) in the Department of Health guidelines.
Participants and methods Participants This study included all 104 general practice groups in Salford & Trafford health district: 236 GPs covering a population of 466 902 (Table 2).
Prescribing Data on vaccine prescriptions during the 1997–1998 ‘influenza season’ (September 1997–January 1998) were derived from the Prescription Pricing Authority prescribing analysis (PACT) data. Data on practice characteristics were derived from Salford & Trafford Health Authority.
Questionnaire The questionnaire was constructed following literature review and piloted in six local practices of varying size, selected to represent areas with a range of socio-economic characteristics. Department of Public Health, University of Liverpool, Whelan Building, Quadrangle, Liverpool L69 3GB. Tim Doran, Specialist Registrar in Public Health Medicine
Introduction The chance of someone susceptible to influenza receiving vaccine depends on which general practice they attend, as responsibility for co-ordinating immunization programmes
Salford & Trafford Health Authority, Peel House, Albert Street, Eccles, Manchester M30 0NJ. Rosemary McCann, Consultant in Communicable Disease Control Address correspondence to Tim Doran. E-mail:
[email protected]
© Faculty of Public Health Medicine 2001
330
JOURNAL OF PUBLIC HEALTH MEDICINE
The final questionnaire was posted to practices, for completion by the person responsible for organizing influenza immunizations. A reminder letter was sent to non-responders 3 weeks later. The questionnaire consisted of: ●
●
●
closed questions to assess which groups of patients were targeted by the practice, and the methods used to identify, contact and vaccinate these patients; nine statements about influenza vaccine and the annual immunization programme, with a five-point rating scale to determine the level of agreement; open questions to determine practitioners’ views on the obstacles to immunization.
Data were analysed and summarized using SPSS (version 6) and Excel (version 5).
Results Prescribing Between September 1997 and January 1998, Salford & Trafford GPs administered 43 572 doses of influenza vaccine, covering 9.3 per cent of the population (Table 2). The proportion of the practice population vaccinated ranged from 0 to 22.6 per cent (median 9.1 per cent, Figure).
Questionnaire Responses were received from 73 of the 104 general practices: a response rate of 70.2 per cent. There were no statistically significant differences between responding and non-responding practices for number of GPs, practice size, number and proportion of patients over age 65, or number and proportion of patients immunized (Table 2).
Targeting patients Table 1 High-risk groups for influenza (1997 Chief Medical Officer’s guidelines) Those with predisposing chronic diseases: pulmonary disease (including asthma) heart disease renal failure diabetes immunosuppression (owing to disease or treatment) Those in residential care and long-stay facilities (e.g. nursing homes) Source (adapted from): UK Health Departments. Immunisation against infectious diseases 1996. London: Department of Health, 1996.
Thirty-six practices (49 per cent) had a policy regarding influenza immunization, but only 25 (34 per cent) had a written policy. Policies were usually based on the Chief Medical Officer’s guidelines (33 of the 36). More than one person was usually responsible for organizing the immunization procedure, with nurses responsible in 64 (88 per cent) of responding practices, practice managers in 46 (63 per cent), and doctors in 31 (42 per cent). Sixty-seven responders (92 per cent) immunized patients during routine patient appointments; practice staff visited nursing and residential homes in 68 cases (93 per cent); and special ‘influenza clinics’ were used by 64 practices (88 per cent), with 1–32 such clinics being set up (median 8). These clinics tended to
Table 2 Characteristics of Salford & Trafford general practice groups
Characteristic General practitioners Total Mean (SD) Median
All practices
Responders (n 73)
Non-responders (n 31)
p value*
236 2.3 (1.4) 2
167 (70.8%) 2.3 (1.4) 2
69 (29.2%) 2.2 (1.4) 2
0.84
466902 4489 (2695) 3737
333175 (71.4%) 4564 (2629) 3999
133727 (28.6%) 4314 (2882) 3052
0.68
Population over age 65 Total Mean (SD) Mean % of practice population Median
71559 688 (438) 15.3% 570
51319 (71.7%) 703 (429) 15.4% 586
20240 (28.3%) 653 (464) 15.1% 515
0.61 0.96
People vaccinated Total Mean (SD) Mean % of practice population Median
43572 419 (281) 9.3% 390
31530 (72.4%) 432 (286) 9.5% 396
12042 (27.6%) 388 (272) 9.0% 329
0.47 0.40
Practice population Total Mean (SD) Median
*Difference in means.
331
OBSTACLES TO INFLUENZA IMMUNIZATION
All the high-risk disease groups specified in the Chief Medical Officer’s guidelines were targeted by most practices, ranging from 83.6 per cent for immunosuppression to 100 per cent for chronic heart disease and chronic respiratory disease (Table 3). Only one of the practices did not target the over-75s, and most practices (61.6 per cent) targeted the over-65s. Other groups not specified in the guidelines were targeted by fewer practices, ranging from 52.1 per cent for carers of high-risk patients to one practice that targeted women in late pregnancy.
be during normal practice hours, Monday to Friday, although 12 practices (16 per cent) organized clinics for Saturday mornings. Nurses administered vaccine in 69 practices (95 per cent), and doctors in 64 (88 per cent). Table 3 Groups targeted by practices and methods used to identify and contact patients Target groups
Number of GPs
% of GPs
Aged 75 Aged 65 Chronic respiratory disease* Chronic heart disease* Asthma* Diabetes* Nursing home residents* Chronic renal disease* Immunosuppression* Carers of high-risk patients NHS employees Students Women in 2nd or 3rd trimester
72 45 73 73 72 72 70 69 61 38 30 9 1
98.6 61.6 100.0 100.0 98.6 98.6 95.9 94.5 83.6 52.1 41.1 12.3 1.4
Methods used to identify patients Practice diabetic register Practice asthma register Computer database Patient notes (paper based) Hospital asthma register Hospital diabetic register Other disease register
65 64 59 44 35 20 9
89.0 87.7 80.8 60.3 47.9 27.4 12.3
Methods used to contact patients Posters Opportunistic contact Reminders on prescriptions Individual letters Newsletters Telephone
71 70 61 29 10 8
97.3 95.9 83.6 39.7 13.7 11.0
Attitudes towards the annual immunization programme Attitudes towards influenza vaccine were generally positive, with respondents perceiving the vaccine to be effective (93 per cent), well received by patients (92 per cent) and without significant side-effects (84 per cent). The annual immunization programme was seen as being necessary (92 per cent), cost-effective (77 per cent), reducing hospital admissions (82 per cent), but very time consuming (64 per cent). There was less agreement about the possibility of a future pandemic, with 8 per cent considering it unlikely, 43 per cent uncertain, and 49 per cent deeming it likely (Table 4).
Obstacles to immunization Fifty-four respondents answered open questions on obstacles to immunization, and four main problem areas emerged: the expense of the annual immunization programme; administrative and practical difficulties; identification of high-risk patients; and patients’ beliefs about influenza and influenza vaccine (Table 5). Financial constraints because of delays between payment of pharmaceutical companies and remuneration by the Health Authority were felt by some practices but not by others, suggesting that arrangements for payment depended on the company and its arrangement with individual practices. The need to anticipate the required quantity of vaccine and order accord-
*Groups specified in the Chief Medical Officer’s guidelines. 16
14
Number of practices
12
10
8
6
4
2
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Percentage of practice population immunized
17
18
19
20
21
22
23
Figure Influenza immunization by Salford & Trafford GPs during winter 1997–1998.
332
JOURNAL OF PUBLIC HEALTH MEDICINE
ingly was also problematic. Estimates tended to be based on the previous year’s demand, and were frequently conservative, as practices did not want to be left with unused vaccine. Although additional vaccine could be acquired if supplies were exhausted early in the campaign, in some cases practices used only the vaccine they had initially ordered. The financial pressure to clear
the stock of bulk-purchased vaccine also led some practices to immunize those people most readily available, rather than those who were most appropriate. This issue was identified by several respondents: ‘Faults we see are with practices offering it to all comers, i.e. worried well, which costs the government without the benefits.’
Table 4 Respondents’ attitudes towards influenza vaccination (numbers, with percentages given in parentheses)
Statement Influenza vaccine is very effective at preventing influenza Influenza vaccine is effective at preventing other respiratory illnesses Annual influenza immunization reduces hospital admissions The annual immunization programme is very time-consuming Influenza vaccine is generally well accepted by patients Influenza vaccine has unacceptable side-effects Annual influenza immunization is unnecessary An influenza pandemic is unlikely to occur Annual influenza immunization is cost-effective
Strongly agree
Agree
Unsure
Disagree
Strongly disagree
16 (21.9) 6 (8.2) 20 (27.4) 11 (15.1) 12 (16.4) 0 (0) 2 (2.7) 2 (2.7) 11 (15.1)
52 (71.2) 26 (35.6) 40 (54.8) 36 (49.3) 55 (75.3) 2 (2.7) 1 (1.4) 4 (5.5) 45 (61.6)
3 (4.1) 17 (23.3) 12 (16.4) 6 (8.2) 2 (2.7) 10 (13.7) 3 (4.1) 31 (42.5) 14 (19.2)
2 (2.7) 19 (26.0) 1 (1.4) 20 (27.4) 4 (5.5) 54 (74.0) 46 (63.0) 32 (43.8) 3 (4.1)
0 (0) 5 (6.8) 0 (0) 0 (0) 0 (0) 7 (9.6) 21 (28.8) 4 (5.5) 0 (0)
Table 5 Obstacles to immunization Obstacle
No. of respondents (n 54)
% of respondents
Expense of the annual immunization programme Delay in remuneration Initial expense of vaccine purchase Insufficient remuneration Wastage of unused vaccine Cost of producing and posting letters Cost of telephone calls
27 12 7 5 4 3
50 22.2 13.0 9.3 7.4 5.6
Administrative and practical difficulties Use of staff time in administering vaccine Use of staff time in preparing for immunizations Interference with other practice work Excessive paperwork/record-keeping Storage of large quantities of vaccine
31 12 12 10 8
57.4 22.2 22.2 18.5 14.8
Identification of high-risk patients Lack of disease registers Need to re-identify patients each year Incompleteness/inadequacy of existing disease registers Lack of clarity/difficulty interpreting Chief Medical Officer guidelines
10 9 3 3
18.5 16.7 5.6 5.6
Patients’ beliefs about influenza and influenza vaccine Vaccine causes influenza or other respiratory disease Influenza is a trivial or self-limiting illness Vaccine does not work Vaccine is required (when inappropriate) Vaccine has unacceptable side-effects Vaccine does not exist
18 12 10 9 4 3
33.3 22.2 18.5 16.7 7.4 5.6
OBSTACLES TO INFLUENZA IMMUNIZATION
There was a consensus that funding additional nursing time would allow immunization clinics to operate without interfering with normal practice business: ‘If money is made available to pay the practice nurse, extra hours could be allotted to run an influenza clinic just during the appropriate season – maybe two hours on a Saturday morning.’ There was, however, disagreement about the potential source of this funding. Although only three respondents disagreed that annual immunization was a cost-effective intervention in NHS terms, seven felt that the immunization programme represented a net cost to the practice each year (‘I doubt whether the cost in time and effort is matched by income from the project’) and that Health Authorities should assist financially. Others saw the programme as an annual source of revenue: ‘Influenza immunization is a money spinner for GPs … vaccine is bought at discount and administered, allowing practices a dispensing fee: 750 vaccines [can be administered] for a profit of £2000.’ Practices identified several ways in which they could be assisted by the Health Authority and/or the Department of Health, including: provision of additional nursing staff/time; provision of additional administrative staff/time; assistance developing ‘influenza registers’; media coverage and publicity; patient education; postal invitations/reminders; item of service payments for administering vaccine; and assistance with supply and storage of vaccine (‘a bigger fridge’). Although several practices felt that additional assistance in the form of postal invitations and media campaigns would be useful, others preferred to continue without interference, either because they considered themselves to be effective enough already or because they did not feel they could cope with additional demand.
Discussion The wide variation in the number of people immunized by practices illustrates some of the problems with an immunization procedure that Watkins criticized for relying on ‘the idiosyncratic behaviour of individuals with minimal central guidance, no mechanisms to ensure effective targeting of vulnerable groups, and no link between remuneration and performance’.7 Patients attending some practices are apparently being neglected, whereas others are being immunized inappropriately. Although the availability of discounted vaccine allows practices to effectively generate a dispensing fee, some consider the personal administration fee inadequate compensation for the expense and disruption of the annual immunization procedure. For others it appears to be an incentive to immunize easy targets, such as students, who fall outside the Chief Medical Officer’s guidelines. Perhaps one reason why practices targeted the over-75s before their inclusion in the Chief Medical Officer’s guidelines is that identification on the basis of age is simpler than identification on the basis of disease. In this case, practices were ahead of UK national policy, and the inclusion of over-75s in the 1998 guidelines may therefore have had little impact. As fewer prac-
333
tices were targeting the 65–74 age group, the introduction of item of service payments for immunizing the over-65s this year may have greater impact, and links remuneration to performance for the first time. Similar incentive schemes in the United States have had some success at increasing vaccine uptake among the elderly.8 The greater rewards may also encourage practices to be bolder with future estimates of vaccine usage and avoid the problem of under-ordering, although pharmaceutical companies could also assist by buying-back unused vaccine. Even if uptake among the over-65s does increase, the new remuneration arrangements may discourage the identification and immunization of younger patients, who are already less likely to be targeted.9 A more difficult proposition, given the variable availability and quality of disease registers, would be to remunerate practices for immunizing all vulnerable groups. Although only 33 practices (45 per cent) had a written policy based on the Chief Medical Officer’s guidelines, the recommended groups were targeted by at least 84 per cent of practices (Table 3). The discrepancy between different high-risk groups suggests some respondents either had incomplete awareness of the recommendations, or that they chose not to target particular groups. Fewer practices targeted groups not recommended in the Chief Medical Officer’s guidelines. Over half targeted carers of people in high-risk groups, often because carers presented with the high-risk recipient and also requested immunization. Carers are mentioned in the guidelines of other countries, such as those of the Centers for Disease Control in the United States,10 and many respondents felt there were sound reasons for vaccinating carers, co-habitants, and (for similar reasons) NHS employees. This year the Chief Medical Officer has begun recommending immunization of NHS and social services workers with direct patient contact. Although students are frequently found in residential accommodation, they are not specifically mentioned in the Chief Medical Officer’s guidelines. The practices that targeted students tended to be those with halls of residence or significant numbers of students within their catchment area. As a group, students are relatively easy to target and vaccinate, as they present to practices in large numbers in autumn to register and receive polio and tetanus boosters. It is questionable, however, whether any genuine benefit is derived from mass immunization of otherwise fit young adults. Previous studies have demonstrated that higher uptake is achieved when physicians have knowledge of guidelines and a positive attitude towards immunization.4,6,11,12 Respondents were generally positive about influenza vaccine and saw a definite benefit in the annual immunization procedure, although 23 per cent had doubts about its cost-effectiveness and 51 per cent were uncertain about the possibility of a future pandemic or thought one was unlikely. It is possible that those responding to the questionnaire were the more enthusiastic practice members, or those with a particular interest in influenza immunization; other practice staff may have been less positive. Conversely, as respondents were those with responsibility for
334
JOURNAL OF PUBLIC HEALTH MEDICINE
organizing the immunization procedure, they bore the greatest administrative burden. As with previous studies, patient beliefs and attitudes were identified as potential obstacles to uptake.5,6,12–14 The best efforts of practice staff can be undermined if there is a lack of public knowledge of, or confidence in, the vaccine. It remains to be seen how successful this year’s nation-wide television campaign will be in encouraging the over-65s to attend for immunization, but such a campaign was requested by many practices. Respondents also requested assistance with postal invitations: one of the more effective methods of increasing uptake.6,15,16 Only 29 respondents (39.7 per cent) adopted this approach, which was seen as time-consuming and expensive. It is an activity, however, particularly suited to Health Authorities, which could easily identify and contact the over-65s, removing a large administrative burden from practices during one of their busiest periods.
Conclusion The annual influenza immunization programme causes considerable disruption to most general practices and financially compromises some. Despite this, most participants in this study recognized its benefits and considered it a necessary activity. Although practices were more likely to target people if the Chief Medical Officer had recommended immunization, prescribing behaviour was not always in line with national policy, adherence to which could be difficult and costly. This year’s guidelines should be both easier to follow and more lucrative. Although many practices were able to conduct successful immunization campaigns without additional funding or assistance, most requested the kind of support the Department of Health is providing this year in the form of advertising campaigns, promotional materials and remuneration. Further co-ordinated support from Health Authorities, pharmaceutical companies and neighbouring practices could help overcome the remaining obstacles to effective immunization.
Acknowledgements We would like to thank all the general practice staff that participated in the study, and Dr Martyn Regan of Liverpool Health Authority for his help and advice. T.D. designed the questionnaire, collated the data, carried out the statistical analyses, and wrote the final version of the paper. R.M. initiated and supervised the study.
References 1 UK Health Departments. Influenza immunisation. London: Department of Health, 2000. 2 Nguyen-Van-Tam JS. Influenza immunisation: policies and practices of general practitioners in England, 1991/92. Health Trends 1993; 25: 101–105. 3 Nicholson KG, Wiselka MJ, May A. Influenza vaccination of the elderly: perceptions and policies of general practitioners and outcome of the 1985–86 immunisation program in Trent, UK. Vaccine 1987; 5: 302–306. 4 Nguyen-Van-Tam JS, Nicholson KG. Influenza immunisations; vaccine offer, request and uptake in high-risk patients during the 1991/2 season. Epidemiol Infect 1993; 111: 347–355. 5 Hak E, Hermens RPMG, van Essen GA, Kuyvenhoven MM, de Melker RA. Population-based prevention of influenza in Dutch general practice. Br J Gen Pract 1997; 47: 363–366. 6 Honkanen PO, Keistinen T, Kivela SL. The impact of vaccination strategy and methods of information on influenza and pneumococcal vaccination coverage in the elderly population. Vaccine 1997; 3: 317–320. 7 Watkins J. Effectiveness of influenza vaccination policy at targeting patients at high risk of complications during winter 1994–5: cross sectional survey. Br Med J 1997; 315: 1069–1070. 8 Kouides RW, Bennett NM, Lewis B, et al. Performance-based reimbursement and influenza immunization rates in the elderly. Am J Prev Med 1998; 14(2): 89–95. 9 Booth LV, Coppin R, Dunleavey J, Smith H. Implementation of influenza immunisation policy in general practice: 1997 to 1998. Commun Dis Publ Hlth 2000; 3(1): 39–42. 10 Centers for Disease Control and Prevention. CDC Surveillance Summaries, April 25 1997. Morbid Mortal Wkly Rep 1997; 46(RR-9): 1–25. 11 van Essen GA, Kuyvenhoven MM, de Melker RA. Compliance with influenza vaccination: its relation with epidemiologic and sociopsychological factors. Arch Fam Med 1997; 6: 157–162. 12 Hershey CO, Karuza J. Delivery of vaccines to adults: correlations with physician knowledge and patient variables. Am J Med Qual 1997; 12(3): 143–150. 13 Lennox IM, Macphee GJA, McAlpine CH, et al. Use of influenza vaccine in long-stay geriatric units. Age Ageing 1990; 19: 169–172. 14 van Essen GA, Kuyvenhoven MM, de Melker RA de. Why do healthy elderly people fail to comply with influenza immunisation? Age Ageing 1997; 26: 275–279. 15 Nexoe J, Kragstrup J, Ronne T. Impact of postal invitations and user fee on influenza vaccination rates among the elderly. Scand J Prim Hlth Care 1997; 15: 109–112. 16 Gill JM. Using mailed patient reminders to increase influenza immunization rates among older adults in a primary care office. Del Med J 1999; 71(10): 427–431.
Accepted on 18 June 2001
