However, using the diagnosis data in combination with drug use as proposed alters the assumption. This is the case particularly when it is possible to assume that the set containing diagnosis data is representative for the population of interest. This diagnosis data could also be used directly to estimate disease prevalence. This algorithm requires a test set with both diagnosis data and drug use. In this paper more specifically the Random Forest (RF) algorithm will be applied, as this method yielded the best results in comparison with others.

Avoiding any a priori assumption on the relationship between the drugs and diagnoses, machine learning algorithms can be used to estimate this relation from data. To overcome these two problems, it is better to estimate the probability of having a specific disease given all different medications a person uses.

Second, some patients with a disease are not prescribed the specific drug, and this will lead to underestimation of prevalence. First, many drugs are used for the treatment of multiple diseases assuming that all patients who take a specific drug do have a specific disease will then lead to overestimation. For two reasons, this procedure is problematic. 9, 13, 14, 18, 19 The links are based on literature or medical guidelines. While drug use data is often recorded without a diagnosis, some studies base disease prevalence estimates on direct links of specific drug use to the presence of certain diseases. Furthermore, GPs may have different coding habits, hindering comparisons between GPs. 15, 16 GP-registers containing diagnosis codes are not readily available in all countries. Hospital data is therefore more likely that GP data to underestimates the prevalence. This means that the GP sees both patients that see only the GP and those he refers to specialist care.

In addition, in the Netherlands, the GP-serves as a gatekeeper, implying that patients-except in emergencies-can only visit a medical specialist with a referral of the GP. Hospital discharge registers are large, but involve hospital-related events only. Surveys are costly to execute on a large scale. 13, 14 In many countries insurers or providers maintain extensive prescription databases, allowing easy access to national drug use data.ĭrug use has several advantages over other sources. 8–12 Drug use data, especially on prescription drugs, has also frequently been used to estimate disease prevalence. 1 Several sources can deliver population disease prevalence estimates, such as surveys, 2–4 dedicated epidemiologic studies using diagnostics 5–7 or administrative data sources. Information on disease prevalence is important for assessing the health needs of populations.

Atc4 code