Nurse telephone triage for same day appointments in general practice: multiple interrupted time series trial of effect on workload and costs Richards D A, Meakins J, Tawfik J, Godfrey L, Dutton E, Richardson G, Russell D Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Nurse telephone triage for same day appointments in the setting of general practice was examined. Receptionists passed on requests for same day appointments to an experienced practice nurse, who managed the call through telephone advice only, a same day nurse appointment, a same day general practitioner (GP) appointment, a home visit, or a routine nurse or GP appointment. Type of intervention Other: Patient care management. Economic study type Cost-effectiveness analysis. Study population The study population comprised patients requesting same day appointments between 8:30 a.m. and 5 p.m. Setting The setting was primary care. The economic study was conducted in York, UK. Dates to which data relate The dates during which the effectiveness and resource use data were gathered were not reported. The price year was not given. Source of effectiveness data The effectiveness evidence was derived from a single study. Link between effectiveness and cost data The costing was conducted prospectively on the same sample of patients as that used in the effectiveness study. Study sample The use of power calculations was not reported. All consecutive patients requesting a same day appointment in the general practice that participated in the study were included in the analysis. Of the 1,263 patients who requested a same day appointment, 1,233 had same day contact under standard management while 30 had no further contact. Of the 3,488 patients who requested a same day appointment under triage management, 3,169 were triaged, 283 were not triaged due to several reasons (e.g. direct call to a GP, out-of-hours calls, health worker referral), and 36 had no further contact. Therefore, the final sample comprised 1,233 patients in the standard management group and 3,452 in the triage Page: 1 / 6
system group. Most of the patients were in the 25- to 44-year age group (27.2% in the standard management group and 25.1% in the triage system group. The proportion of female patients was 61.1% in the standard management group and 60.6% in the triage group. Study design The authors reported that this study was of a multiple, interrupted time series design. The study practice, which was in York (UK), had three participating surgeries and a study population of 20,800 patients. Data from each surgery were gathered for one week per month over a 12-month period. The first three repeated measurements were carried out to assess standard management. After the introduction of the new triage system, the subsequent nine repeated measurements were carried out to assess triage management. Surgery sites entered the study sequentially at three monthly intervals. The study design was selected so as to control for factors such as changes in professional expertise, or health or epidemiological trends. Autocorrelation analysis and control for confounding factors were also carried out. Both systems had been pilot tested before implementation. The patients were followed-up for one month after the study appointment. No loss to follow-up was reported. Analysis of effectiveness The analysis of the clinical study was conducted on an intention to treat basis. The primary outcome measures used were: the final point of contact (whether nurse phone, nurse appointment, GP phone, home visit, or GP appointment); nurse, GP, and total time per patient; and follow-up care. Follow-up care was assessed as the proportion of patients who returned for further practice-based care within one month (duration of appointment was also recorded), and the number of out-of-hours consultations, accident and emergency visits, and return consultations. The study groups were comparable at baseline, but the triage group had significantly more patients presenting complaints related to the respiratory and dermatological system than patients in the standard management group. The triage group also had fewer mental health complaints. Time series analysis, analysis of variance and regression models were used to assess the impact of autocorrelations or seasonal effects, and potential prognostic factors. Relative risks (RRs) and confidence intervals (CIs) were calculated. Effectiveness results The final point of contact was: nurse phone for 1.8% of the standard group patients and 25.8% of the triage group patients, nurse appointment for 8.7% (standard group) and 14.9% (triage group), respectively, GP phone for 11.3% (standard group) and 6.5% (triage group), home visit for 14.3% (standard group) and 12.1% (triage group), and GP appointment for 64% (standard group) and 40.8% (triage group). Therefore, the triage system led to fewer GP visits, but resulted in more telephone consultations (RR 2.41, 95% CI: 2.08-2.80) or nurse care (RR 3.79, 95% CI: 3.21-4.48). A small reduction in home visits was observed (RR 0.85, 95% CI: 0.72-1.00). Page: 2 / 6
The statistical analyses showed that there was no significant autocorrelation or seasonal effect on the results of the analysis. A learning curve effect after triage was not observed. The mean nursing time per patient was 1.02 (+/- 3.09) minutes in the standard group and 5.17 (+/- 3.83) minutes in the triage group, (p<0.001). The mean GP time per patient was 8.65 (+/- 7.35) minutes in the standard group and 6.20 (+/- 7.56) minutes in the triage group, (p<0.05). The mean total time (GP plus nurse) per patient was 9.67 (+/- 6.98) minutes in the standard group and 11.37 (+/- 7.42) minutes in the triage group, (p<0.001). The regression analysis revealed that nurse time was influenced by the management system. GP time was affected by the number of patients at each final contact point. In relation to follow-up care, more patients in the triage system returned for further practice-based care within one month (RR 1.11, 95% CI: 1.01-1.22). The average duration of appointment was 9.67 (+/- 6.98) minutes with standard management and 11.37 (+/- 7.42) minutes with the triage system. The mean difference was 1.70 minutes (95% CI: 1.24-2.16; p<0.001). The mean number of out-of-hours consultations was 0.08 (+/- 0.38) in the standard management group versus 0.11 (+/- 0.49) in the triage group. The mean difference was 0.04 (95% CI: 0.01-0.07; p=0.005). The mean number of accident and emergency visits was 0.010 (+/- 0.10) in the standard group versus 0.033 (+/- 0.19) in the triage group. The mean difference was 0.023 (95% CI: 0.015-0.032; p<0.001). The mean number of return consultations was 0.93 (+/- 1.30) in the standard group versus 1.24 (+/- 1.78) in the triage group. The mean difference was 0.32 (95% CI: 0.22-0.41; p <0.001). Statistical tests showed that patients were more likely to have contact with the practice after same day telephone care than after appointments (RR 1.32, 95% CI: 1.23-1.41), and after nurse care rather than after GP care (RR 1.15, 95% CI: 1.08-1.23). Clinical conclusions The effectiveness analysis showed that the triage system reduced GP workload, but the mean total time per patient remained unaltered due to the higher number of patients attended by nurses. In addition, more patients in the triage system returned for further practice-based care within one month than did patients in the standard management group. Measure of benefits used in the economic analysis The health outcomes were left disaggregated and no summary benefit measure was used in the economic analysis. In effect, a cost-consequences analysis was conducted. Direct costs Discounting was not relevant since the costs were incurred during a short time. The unit costs and the resource quantities were not presented separately. The health services included in the economic evaluation were GP and nurse time, drugs, tests and emergency care. The cost/resource boundary of the study was not totally clear, but it appears to have been that of the UK NHS. Individualised resource use data were gathered prospectively from the same sample of patients as that used in the effectiveness study. Average time was used for follow-up telephone consultations, appointments, or home visits, as they were not timed. The costs were estimated from salary and earning scales, the British National Formulary, and local provider units. The price year was not reported. Statistical analysis of costs The costs were presenting as mean values with standard deviations. An independent t-test was used to test the statistical Page: 3 / 6
significance of differences in the costs estimated in each group. Bootstrapped estimates were also calculated due to the skewed distribution of the costs. Indirect Costs The indirect costs were not considered. Currency UK pounds sterling (). Sensitivity analysis Univariate sensitivity analyses were conducted to address the issue of variability in the data. The total costs were recalculated using lower and higher estimates of staff costs, and both local and national estimates of emergency and outof-hours costs. Further details on the type of sensitivity analysis were not provided. Estimated benefits used in the economic analysis See the 'Effectiveness Results' section. Cost results The mean total costs per patient were 21.89 (+/- 23.89) in the standard group and 23.37 (+/- 30.65) in the triage group. The difference was 1.48 (95% CI: -0.19-3.15; p=0.81). GP same day appointment costs and drug costs were significantly lower in the triage group. However, nurse same day appointment costs, nurse follow-up costs, and out-of-hours and accident and emergency costs were significantly higher in the triage group. Overall, the difference in total costs did not reach statistical significance. The bootstrapped 95% CI for the total costs was -0.23 to 3.02. Synthesis of costs and benefits The costs and benefits were not synthesised. The sensitivity analysis showed that the cost-difference (mean cost per patient in the triage group minus mean cost per patient in the standard group) ranged from 0.71 (95% CI: -0.69 to 2.11; p=0.32) with lower cost estimates to 2.32 (95% CI: 0.42-4.22; p=0.017) with higher cost estimates. When the accident and emergency costs were removed, the costdifference was not statistically significant. Authors' conclusions The implementation of a triage system reduced general practitioner (GP) same day appointments. However, more triaged patients returned to surgery within one month after triage than standard management patients. Moreover, no costsavings were observed. CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparators was clear. Standard management of patients requiring same day appointments was appropriate, as this reflected the routine approach in the authors' setting. A description of the standard approach was given. You should decide whether this is a valid comparator in your own setting. Page: 4 / 6
Validity of estimate of measure of effectiveness The analysis of effectiveness was based on study with a quasi-experimental design, which was appropriate for the study question. The study had several strengths. First, intention to treat was the basis for the analysis of the clinical study. Second, the large sample size. Third, the use of statistical tests to assess the issues of autocorrelation, impact of confounding factors, and other problems such as confounding factors. Fourth, the length of follow-up was appropriate. Finally, the effectiveness data were derived from multiple centres. The patient sample appears to have been representative of the study population. However, no explicit justification was provided for the choice of the sample size, and power calculations were not reported. In addition, the study groups were not totally comparable at baseline, as the typology of complaints differed between the groups. Validity of estimate of measure of benefit No summary benefit measure was used in the analysis because a cost-consequences analysis was conducted. Validity of estimate of costs The authors did not explicitly report the perspective adopted in the study. Hence, it was unclear whether all the relevant categories of costs had been included. Overall, it appears that the perspective of the health care provider has been adopted. The source of the cost data was reported, but information on the price year and unit costs was not. This would make both reflation exercises and replication of the study in other settings difficult. The authors stated that the baseline difference in study groups, in terms of the type of presenting problems, did not affect the cost analysis. Appropriate statistical tests were used to address the issue of non-normal distribution of the costs. Some cost items were varied in the sensitivity analysis. Other issues The authors made some comparisons of their findings with those from other studies. They found that their results were consistent with those from one published study on nurse triage, but discrepancies with another study were observed. The authors noted that the differences could have been due to the much longer follow-up in the present study. The advantages and disadvantages of the new triage system were highlighted. The issue of the generalisability of the study results to other settings was not addressed and sensitivity analyses were conducted only for a limited number of cost estimates. Caution is therefore required when extrapolating the study results to other settings. Implications of the study The authors noted that future studies should address specific issues, such as whether the additional use of out-of-hours or accident and emergency services was a consequence of patients not having their needs meet. Future studies should also address the impact of the intervention on health and quality of life. Source of funding Supported by a grant from the NHS Executive Northern and Yorkshire Regional Office Responsive Funding Programme. Bibliographic details Richards D A, Meakins J, Tawfik J, Godfrey L, Dutton E, Richardson G, Russell D. Nurse telephone triage for same day appointments in general practice: multiple interrupted time series trial of effect on workload and costs. BMJ 2002; 325: 1214-1217 PubMedID 12446539 Other publications of related interest Gallagher M, Huddart T, Henderson B. Telephone triage of acute illness by a practice nurse in general practice: Page: 5 / 6
Powered by TCPDF (www.tcpdf.org) outcomes of care. British Journal of General Practice 1998;48:1141-5. Lattimer V, George S, Thompson F, et al. Safety and effectiveness of nurse telephone consultation in out of hours primary care: a randomised controlled trial. BMJ 1998;317:1054-9. Indexing Status Subject indexing assigned by NLM MeSH Appointments and Schedules; Costs and Cost Analysis; Economics, Nursing; England; Family Practice /economics /organization & administration; Humans; Nursing Care /organization & administration; Patient Acceptance of Health Care /statistics & numerical data; Telephone; Time Factors; Triage /economics /organization & administration; Workload AccessionNumber 22002008296 Date bibliographic record published 30/11/2004 Date abstract record published 30/11/2004 Page: 6 / 6