Information given to patients before appointments and its effect on non-attendance rate

Background

Outpatient non-attendance is a common source of inefficiency in healthcare provision, wasting time and resources and potentially lengthening outpatient waiting times. Non-attendance occurs in all age groups and in people from various different social, cultural, and ethnic backgrounds; it affects all specialties and does not seem to be restricted to a particular healthcare sector.1^–3

Strategy for change

As previously, on receipt of a referral letter, patients were sent an appointment some three to four months before their appointment. Two weeks before this appointment patients were sent detailed information outlining their forthcoming visit, and one week before the appointment they were phoned.

The information pack was written in large type and in simple language. The first page stated the name of the clinic, the doctor's name, the precise location of the clinic, the day, date, and time of the appointment, where to park, and what to bring. The second page described exactly what would happen to the patient, from a greeting by the clinic nurse to blood tests and departure. Page three explained what would happen after the appointment during a structured diabetes education programme over five weeks, again specifying who they would see, when, and where, and then similar details for the discharge assessment clinic and the discharge process. Page four stated “our aims,” a mission statement, and specified what number to phone if they are unable to attend; and page five invited positive or negative feedback about our service and was accompanied by a stamped addressed envelope directed to the general manager for medicine. The final page was a flow diagram summarising each step of the patient's visit with boxes to be ticked by hospital staff as each step in the process was completed.

The phone call one week before their appointment confirmed receipt of the pack and their intention to attend their appointment.

Process of gathering information and key measures

We compared non-attendance rates after our intervention with rates in the same clinic over the preceding two to three years to try to ensure that any change after our intervention was not a result of some underlying temporal trend in non-attendance rates in this clinic. In addition, we also examined non-attendance over the intervention period in another clinic to establish that any changes in our clinic were not the result of some more generalised change affecting all clinics within the hospital.

Non-attendance rates were expressed as percentage of total appointments (attenders plus non-attenders plus cancellations) and were compared by χ² tests with Yates's correction.

Effects of change

In 325 consecutive patients, all of whom were sent the patient information pack, 147 (45%) received information pack plus follow up telephone call and 178 (55%) received information pack without telephone call. Patients did not receive a follow up call if we were unable to obtain a telephone number from the general practitioner or the hospital information system, if they were known not to have a telephone, or if after repeated calls (typically three or four) the secretary was unable to contact them. The historical control group comprised 1336 patients who had been sent appointments for the same clinic in the two to three years before the study. None of these patients had received an information pack or phone call.

Overall, the information pack was associated with a significant reduction in non-attendance. In new patients who were given information before their appointment (with or without phone call) 4.6% (15/325) did not turn up compared with 15% (201/1336) of those who had received neither pack nor phone call (P<0.0001). Of the new patients who received both information pack and phone call, 1.4% (2/147) did not attend compared with 7.3% (13/178) who received information but no phone call (P<0.01).

Lessons learnt

Outpatient non-attendance is common, wastes time and resources, and lengthens waiting times. Reducing non-attendance offers an opportunity to make better use of healthcare resources and to reduce waiting times.

Our study shows that fully informing patients about their appointment dramatically reduced outpatient non-attendance. Non-attendance rates of 15-19% may be reduced to about 7% by sending patients information. Moreover, a further reduction in non-attendance to about 1% may be achieved by following up information with a telephone call one week before the patient's appointment.

Non-attendance in our historical control group was high (15-19%) and remarkably similar to the rate in our hospital as a whole and to that of a neighbouring university hospital (both about 15%). This suggests that non-attendance in our clinic before the information pack was representative of non-attendance in other clinics and in other institutions. The non-attendance rate over the preceding three years in this clinic was 9%, 13%, and 19% for the years 1996-8, which suggests that the dramatic fall after introduction of our information pack and phone call was unlikely to be a result of a background trend in non-attendance rates. Moreover, although our diabetes review clinic catered for a different group of patients (follow up patients), non-attendance rates were similar to those in the new patient clinic, the hospital as a whole, and the neighbouring university hospital before the study and showed no sign of decreasing over the study period. This suggests that the fall in non-attendance in the new patient clinic was likely to have been as a result of the information pack and phone call and not the result of some coincidental process favourably affecting non-attendance in all the hospital's outpatient clinics.

It is possible that the smaller reduction in non-attendance seen in patients who received the information pack but were not phoned was not a consequence of not being phoned but of some other factor associated with not having a telephone, such as poverty. This is unlikely as only a minority of patients who were not phoned did not have a telephone.

Potamitis et al found that over a quarter of non-attendance could be attributed to clerical error and nearly a fifth to patients forgetting appointments and suggested that a simple postal reminder could prevent 40% of non-attendance.4 In practice, however, Majeed and colleagues found that reminder letters were of only limited benefit.5 Livianos-Aldana et al found that the shorter the time interval between the appointment letter and the appointment, the lower the non-attendance rate.6 Like us, Seow et al found that personal contact together with an information pack reduced non-attendance, but in their study an information pack alone was not effective.7 It would seem therefore that timing and content of material sent to patients is important and that personal contact increases the benefit obtained from an information pack alone.

In addition to several hours of secretarial time, there were some minor additional costs for materials and phone calls associated with our project. Anecdotally, however, these seem to have been more than offset by reductions in the number of letters concerning non-attendance, cancellations, rescheduled appointments, patient and general practitioner follow up calls about missed appointments, and, most importantly, more effective use of existing appointment slots.