Quality indicators in headache care: an implementation study in six Italian specialist-care centres

Of more than 80 headache specialist centres in Italy, ten participate in a network (the Young Italian Headache Network) of young Italian researchers on headache; seven of these expressed their willingness to participate in the study. However, one centre discontinued its participation for unknown reasons, without contributing any data. The study was completed over 3 months, between September and December 2016, in the remaining six centres (Table 1). The centre of Roma Sapienza, involved also in the earlier study of Schramm et al. [7], collected new data in accordance with our study protocol.

In each centre we collected data from service records (via the service manager), from the physicians and other HCPs (nurses, psychologists and/or nursing assistants) involved in outpatient consultations, and from 60 consecutive patients (30 enrolled when making a first-time visit and 30 others undergoing a follow-up visit).

We employed the five evaluation instruments used by Schramm et al. [7]. Table 2, set out similarly as in the earlier studies [6, 7], provides an overview of these. They included three different questionnaires, one each for the physicians, other HCPs and patients, all completed anonymously. The questionnaire for patients was formulated as an exit questionnaire, which they received from their HCPs at the end of their consultations and completed and returned before leaving. All questionnaires were translated from their English originals into Italian by two authors (LP and SG) according to Lifting The Burden’s translation protocol for hybrid documents [9], with TJS and physicians and patients not involved in the study providing review and assistance.

In addition, data were extracted into two data-collection forms from the patients’ medical records and from service records. This was done by a local investigator at each centre who, to avoid any bias, did not perform any consultations with patients enrolled in the study.

Data were entered into spreadsheets at each centre, and then transferred to the data collection centre (Headache and Drug Abuse Centre, University of Modena and Reggio Emilia).

Descriptive analysis was performed by LP. Demographic and clinical data were provided as numerical values and are summarized as percentages or mean values. No hyphotheses were statistically tested.

Diagnoses were recorded using ICHD terminology in ≥97% of cases. All diagnoses at every centre were documented after the first visit and reviewed during follow-up using diagnostic diaries. Temporal duration of the presenting headache disorder was reported in ≥88% of patients’ records (Table 3).

Waiting-list times for first appointment were not documented, but all clinics reported a formal triage system to expedite appointments in case of perceived urgency.

Mean time allocated to patients’ visits ranged, according to patients’ reports, between 26 (Padova) and 49 min (Bologna) for the first visit and between 17 (Firenze) and 29 min (Bologna and Roma Campus Bio-Medico) for follow-up visits. Patients’ and HCPs’ satisfaction with time per visit was ≥83% in all centres.

Frequency of symptoms and treatment plans were properly reported in most cases (respectively >82 and >87%). Psychological therapies were available in four of the six centres: in two of these (Roma Campus Bio-Medico and Roma Sapienza) their prescription was correctly reported in patients’ records, while in the other two (Padova and Firenze) these therapies were not prescribed to any patients enrolled in the study. Disability instruments were available in three centres (Padova, Modena and Firenze), but no HCP utilized them regularly. Follow-up calendars were available and recorded future appointment dates in all centres (Table 3).

Referral pathways were well established in most centres: exceptions were Roma Campus Bio-Medico for primary care and Padova for urgent consultations (Table 3).

Most patients (> 85%) expressed satisfaction with the information given by physicians, and >82% declared they had received appropriate reassurance (Table 3).

Both patients and HCPs found the service environment clean and comfortable in four centres, with some dissatisfaction expressed in two (Bologna and Firenze). Patients’ satisfaction was generally high with the welcome they received from the services (67–98%) and with waiting times to be seen in the centres (67–82%). Average waiting times varied little, 8–15 min for first visits and 5–25 min for follow-up visits, and were unsatisfactory only for the HCPs in one centre (Bologna) (Table 3).

Overall satisfaction with their management was expressed by 78–95% of patients. (Table 3).

All clinics had a protocol to avoid wastage of resources, and input costs were regularly recorded, but not all physicians were aware of these in one centre (Firenze) while the information was available to senior management only in another (Padova). All centres provided equal access for all who might need it.

Outcome assessments were made routinely only in Padova, and were based on symptom burden. Neither disability nor quality of life were assessed routinely at any centre.

All services appeared safe: prescribed drugs were well documented, protocols for reporting serious adverse events (SAEs) existed in each centre and there were no recorded SAEs.

The consensus group developing these indicators foresaw their employment in benchmarking, but deferred exploration of this potential important application [5]. Our findings in these six centres suggest that the indicators may indeed be deployed in setting benchmarks – at least in Italian specialist headache care. They clearly identified majority practice, with some aspects requiring improvement. Majority practice fails to make routine use of formal outcome measures, despite that this is an obvious part good record-keeping, and would appear to be necessary to judge whether best outcomes are achieved. The evident practical (lack-of-time) difficulty in doing this needs to be addressed. It should be recognized that HURT, for example, is intended for self-completion [13, 14], and can be brought by patients to follow-up visits already filled in – and not only used for its purpose of guiding treatment (less necessary in specialist care) but also placed in the patients’ records. The solution may lie in developing an electronic version, which has not so far been done. In the context of benchmarking, there are aspects of care not currently included in the quality indicators that many might think should be, and which at least deserve to be discussed for the future. One that is clearly relevant is whether diagnostic examinations and the prescriptions of treatments follow evidence-based guidelines [15, 16]. The consensus group responsible for developing the indicators acknowledged this, but also recognized that assessment of these was highly problematic [5]. A second aspect, not included in the validated set of quality indicators but perhaps important nonetheless, is the non-medical barriers to care erected, before visits, by various impositions on patients (finding and paying for parking space, for example). We asked patients about their satisfaction with regard to this particular aspect, in a similar manner to the other quality indicators (“yes”/“not yes” responses), obtaining positive replies as follows: Firenze 45%, Roma Sapienza 53%, Modena 68%, Padova 80%, Bologna 82%, Roma Campus Bio-Medico 87%. Although not a validated quality indicator, this (or an adaption of it) might identify additional difficulties encountered by patients in receiving medical care from the specialist centres that are relevant to service quality.