General practice patients in treatment for substance use problems at different levels of care: an observational study using two data sources

Background

In Belgium, the incidence of treated substance use problems is monitored by the Network of Sentinel General Practices (SGP) surveillance at the primary care level and by the Treatment Demand Indicator (TDI) surveillance at higher, more specialized care levels. We examined whether the TDI-population that was referred by a GP to have specialized treatment is similar to the SGP population treated by a GP for substance use problems.

Methods

The TDI-surveillance is based on data of new treatment episodes reported by a large range of specialized care services. The SGP surveillance is based on all new episodes in a representative sample of general practices. Both surveillance systems have a common core of variables defined by a European protocol. Using TDI-data (N=60,310) from 2016-7, we compared characteristics of patients that were referred by a GP (“GP-referred”) to patients referred by other care services, after exclusion of self-referred patients or patients “encouraged” by persons or services at the preliminary care level. Next, the GP-referred TDI-population was compared to the population that was treated by their GP (“GP-treated”) for a new episode of substance use problems, using SGP-data (n=314) from the same period. Data were examined using 95% confidence intervals (CI) and multivariate regression.

Results

According to TDI-data, GP-referred patients (n=4,522) distinguish themselves from patients referred by non-GP care services (n=12,054) by higher age (OR 1.14; 95% CI 1.03-1.26), higher education (above primary level) (OR 1.19; 95% CI 1.07-1.32), first substance use treatment episode ever (OR 1.64; 95% CI 1.50-1.79), no cannabis problems (OR 0.73; 95% CI 0.63-0.85), mono-substance use (OR 1.14; 95% CI 1.03-1.26), recent non-abstinence of primary substance (OR 1.93; 95% CI 1.64-2.27), having a job (OR 1.69; 95% CI 1.53-1.86) and a fixed residence (OR 2.97; 95% CI 2.52-3.51). GP-referred patients were relatively more likely to receive care in (a psychiatric service of) a general hospital (OR 1.85; 95% CI 1.65-2.07) and less likely in a short-term or long-term residential care setting outside general and psychiatric hospitals.

A comparison between the GP-referred population (TDI) and the GP-treated population (SGP) showed similar distributions of gender, age, treatment history, age of first substance use and recent (non-) abstinence. Adjusted odds showed that GP-treated patients were characterized by previous substance use treatment (OR 1.56; 95% CI 1.19-2.03), problems with prescription drugs (mainly hypnotics and sedatives) (OR 1.83; 95% CI 1.24-2.70), mono substance use (OR 1.58; 95% CI 1.15-2.16) and recent professional activity (OR 2.00; 95% CI 1.55-2.59).

Conclusions

Patients with a new treatment episode for substance use problems treated by their GP and GP-referred patients treated on higher specialized care levels are highly comparable. Yet, the SGP reported more previous treatment of their patients than TDI-caregivers, probably because the latter do not consider GP care of patients with substance use problems as treatment. The SGP also reported more problems of prescription drugs among their patients compared to the GP-referred TDI-population. A possible explanation is that relatively less GP-treated patients with problems of abuse or misuse of prescription drugs seek specialized care but the accessibility of appropriate care may pose equally problems for these (older) patients.