Every day, in Catalonia, thousands of people visit one of its 371 public primary healthcare centres and many of them come away with a prescription to start a treatment. However, in a study of 1.6 million patients and 2.9 million prescriptions carried out between July 2013 and July 2014, it was observed that 17.6% of medications prescribed for the first time were never collected. According to the results, patients who do not initiate treatment spend less on medication but their economic impact on society is greater due to longer sick leave and a greater use of emergency healthcare services. 

The study shows that non-initiation is more frequent among young men, among people originating from the Americas and in those requiring symptomatic or acute treatment. Being seen by a doctor who was not their regular doctor or at a healthcare centre that trains junior doctors also increased the possibilities of patients not initiating their treatment.

Thirteen categories of medicines were studied, including the ten most prescribed groups in Catalonia and the five most expensive. These include antibiotics, painkillers, antihypertensives and antidepressants, among others. If we focus on medication types, we observe that for painkillers, one out of every five prescriptions is not collected, whereas in hypertension or heart failure treatment prescriptions the percentage is notably reduced, as these are considered by patients to be severe conditions.  

However, it is among medications for treating chronic health problems, such as diabetes or depression, where the most alarming data are found. One out of every ten prescriptions for insulin or antidepressants were never collected from the pharmacy. 

The study also shows that the majority of patients willing to initiate the treatment prescribed do so during the first month, as a difference of only 1.5 percentage points is observed between the number of prescriptions not collected in the first month and those not collected in the following two months. Similarly, it is perceived that once three months have elapsed since the treatment was prescribed, the number of chronic patients who have not initiated their treatment decreases, while some of those who were prescribed painkillers continue without collecting their medications.

This latter case can be explained by the so-called medicine-cabinet effect, in other words, patients do not pick up their prescription because they either already have medications from previous prescriptions or can find them easily at home because they have been prescribed for other family members. Another reason may be the fact that these types of medications can be freely purchased over the counter at pharmacies, which makes it easier for patients to obtain them without leaving any record of having initiated their treatment.

Young malesare those who least tend to initiate the pharmacological treatment prescribed by their doctor, whereas women and very elderly people are those who initiate treatment most. The age differences may be due to young people not perceiving their illness as a risk or reacting to a diagnosis by refusing to undergo the treatment advised by their doctor.

However, gender and age are not the only factors that influence the typification of patients who do not initiate medication. Nationality is also a variable that must be taken into account. The study reveals that patients from the American continent take less notice of doctors’ indications, which could be due to misconceptions and a lack of confidence in the functioning of the public health system in their host country. Meanwhile, the patients who most follow the advice of practitioners are those of African or Asian origin, while there are major differences among those suffering from mental illness or who have a disability. These latter two types of patients have, in general, a greater risk for not initiating the treatments prescribed for them. Furthermore, non-initiation is even more frequent if patients in this group are African or Asian. This is probably due to cultural differences in explaining their symptoms or the stigma associated with them. 

The prevalence for non-initiation of a prescribed treatment changes with increasing age among patients of different nationalities. Generally, a higher proportion of elderly people usually start their medication. If we focus specifically on their geographical origin, the population of American origin has the highest possibilities of not initiating treatments, whereas the African population initiate them most. People of Spanish origin are more inclined towards not taking the medications prescribed for them than Asians, especially on the younger population. If we look at the older population, this situation changes. Previous research on access to health services among migrants suggests that this occurs thanks to better adaptation to the health system and culture.

In comparison with other nationalities, young American and Spanish people are those who tend most not to initiate treatments, whereas elderly Spanish people, together with Africans, are those who most follow the doctor’s prescription from the start. More in-depth research into patient beliefs regarding medication and health conditions would enable the production of more comprehensive patient profiles, a task that is pending to date.

For the construction of standard patient profiles, the data used are sourced from the computerised system of the Catalan Health Institute (ICS) and the Information System for the Development of Primary Healthcare Research (SIDIAP). These data are completely anonymised and under no circumstances enable identification of any individual patient. 

Patient decisions to collect prescribed medication from the pharmacy and start taking it also depends largely on their relationship and degree of trust towards their general practitioner. When faced with a new prescription, patients evaluate the implications and risks of the diagnosis, the doctor’s indications, the type of treatment proposed and what benefits they may gain from the medication. In this phase, patients’ trust in their physician is a key factor when deciding whether to begin treatment or not.

From an analysis of profiles of nearly 7,000 family doctors at Catalan public healthcare centres, the conclusion is drawn that when the prescribing practitioner is a locum doctor, there is an increased possibility of the patient not initiating treatment. When not treated by their regular practitioner, patient-doctor trust is considerably undermined, which raises more doubts among patients regarding diagnosis and indications.

The type of healthcare centre in which the prescription is issued also determines the number of prescriptions that will be collected from the pharmacy. According to the study conducted, prescriptions made out at teaching centres (healthcare centres where junior doctors are trained) are used less. This could be because at these types of centres, the practitioner attending the patient is a junior doctor, in other words, a professional who is taking a foundation programme (MIR). A lack of confidence in the knowledge and experience of doctors still in training could be the main cause of this behaviour.

The results of the study indicate that it would be beneficial to design specific initiatives to strengthen patient trust in junior and locum doctors. It is also important for doctors to resolve any doubts patients may have regarding their diagnosis and prescribed treatment, providing accurate facts to help them make an informed decision on whether to initiate medication.

Although many studies have been carried out on the social impact of non-continuation of treatments already begun, to date there has been no study of the impact of treatments that never reach initiation. For this purpose, the costs of medical services and medications supplied are taken into account along with days of sick leave. These are converted into losses in productivity and costs for society. The wealth of data available on the public health system in Catalonia have enabled this analysis. 

According to the results, patients who do not initiate treatment spend less on medication, but their economic impact on society is greater due to an increased duration of sick leave and greater use of emergency healthcare services such as home visits. The study shows that patients who decide not to start taking the medication prescribed for them visit their healthcare centre less frequently and undergo fewer medical tests, which means they generate a lower healthcare spend. However, these types of patients require more home visits by doctors or nurses, which is a more expensive service than appointments at healthcare centres.

The use of these services and the fact that they spend more days on sick leave also indicate a worsening of patients’ state of health, which could ultimately lead to longer treatments that, probably, will result in higher consumption of medical resources and more days on sick leave.

In terms of losses in productivity due to sick leave, patients who do not initiate medication spend more days off sick on average, with a mean difference of 2.5 days. Therefore, expense related to incapacity for work has to be added to the economic losses of the healthcare system due to the use of more expensive services. This costs are assumed by companies, mutual societies and the healthcare system, among others. 

When the use of these services and sick leave are translated into economic figures, it is deduced that patients who do not initiate a treatment prescribed by a primary healthcare doctor are responsible for greater spending in comparison with initiator patients. On average, the extra cost is between 3 and 165 euros per patient, depending on whether the minimum wage or the average wage is used to calculate the losses caused by sick leave. In Catalonia, taking the average wage as a reference, the accumulated excess cost of all non-initiator patients may reach up to 44.5 million euros per year.

The consequences of non-initiation vary according to the treatments prescribed. While the effect of non-initiation of treatments with antibiotics, painkillers or antidepressants is clearly visible in the short term, due to a deterioration in patients’ health – and consequently more days of sick leave – in chronic illnesses, such as hypertension, it has not been possible to evaluate the impact. The complications of non-initiation of treatment in these types of illnesses may take a longer time to appear, between five and ten years, therefore the monitoring of patients over two years is not conclusive. It is for this reason that, when taking into account the minimum wage to estimate the impact on costs of non-initiation of chronic treatments, non-initiator patients generate lower spending than initiators, whereas with the average wage no differences are observed.

This study was not able to evaluate the use of hospital services, such as admissions, treatments and emergency visits. If the relevant data were taken into account, the economic cost of patients deciding not to initiate treatment could be even greater. In the sphere of productivity losses, the impact could also be greater if data were available to assess the effect of non-initiation of treatment among the population without any employment contract, such as many self-employed people, and among the non-earning population, such as students or retired people.

In this study, it was assumed that patients who collect a first prescription are going to initiate that treatment. Although this is not going to be true in every single case, it will be in the majority, as has been shown in previous research.

Similarly, the costs derived from hospital services have not been taken into account. These are higher than those generated by primary healthcare, and could modify the results shown in terms of costs of non-initiation of medication in some way. However, the explorations carried out seem to indicate that, even considering these costs, non-initiators would continue to generate greater spending than initiators.

This study enables a better understanding of the scale of both the frequency and the economic impact of the decision to not initiate a medical treatment. It also enables an understanding of which types of patients are more likely not to start treatment, although more in-depth research is needed to understand the reasons for this behaviour.

This information is considered key for designing new strategies and interventions, with the objective of minimizing the number of patients that, whether due to lack of confidence in the prescribing doctor, socioeconomic reasons, or their own perception of the illness diagnosed, decide not to take the medication prescribed for them. The application of specific policies and interventions to reduce this behaviour should be implemented as soon as possible in the primary healthcare system in Spain and requires the collaboration of all stakeholders involved, including doctors, nurses, social workers, specialists and pharmacists. 

Among steps to be taken in this area are the conducting of new studies that assess the long-term consequences, in terms of costs and health, of non-initiation of treatment in chronic illnesses, as well as a study of the social and economic impact of this behaviour among the population of children and young people.

This article has been adapted from these studies:

AZNAR-LOU, I. A. FERNÁNDEZ, M. GIL-GIRBAU, M. FAJÓ-PASCUAL, P. MORENO-PERAL, M.T. PEÑARRUBIA-MARÍA, A. SERRANO-BLANCO, A. SÁNCHEZ-NIUBÓ, M.A. MARCH-PUJOL, A.M. JOVÉ and M. RUBIO-VALERA (2017): "Initial medication non-adherence: prevalence and predictive factors in a cohort of 1.6 million primary care patients", British Journal of Clinical Pharmacology, 83(6):1328-1340.

AZNAR-LOU, I. A. FERNÁNDEZ, M. GIL-GIRBAU, R. SABÉS-FIGUERA, M. FAJÓ-PASCUAL, M.T. PEÑARRUBIA-MARÍA, A. SERRANO-BLANCO, P. MORENO-PERAL A. SÁNCHEZ-NIUBÓ, M.A. MARCH-PUJOL and M. RUBIO-VALERA (2017): "Impact of initial medication non-adherence on use of healthcare services and sick leave: a longitudinal study in a large primary care cohort in Spain", British Journal of General Practice, 67(662):e614-e622.

AZNAR-LOU, I., M IGLESIAS-GONZÁLEZ, M. GIL-GIRBAU,  SERRANO-BLANCO, A. FERNÁNDEZ, M.T. PEÑARRUBIA-MARÍA, R. SABÉS-FIGUERA, A.G. MURRUGAS-CENTURIÓN, M. MARCH-PUJOL, M- BOLÍVAR-PRADOS and M. RUBIO-VALERA(2018): "Impact of initial medication non-adherence to SSRIs on medical visits and sick leaves", Journal of Affective Disorders,226:282-286.

More bibliographic references:

BOLÍBAR, B.F. AVILÉS, R. MORROS R,M. GARCÍA-GIL. HERMOSILLA, R. RAMOS, M. ROSELL, J.RODRÍGUEZ, M. MEDINA, S. CALERO and D. PRIETO-ALHAMBRA (2012): "SIDIAP database: electronic clinical records in primary care as a source of information for epidemiologic research", Barcelona, Medicina Clínica, 138(14):617-21.

HUTCHINS, D.S., J. E. ZEBER, C.S ROBERTS, A.F., WILLIAMS, E. MANIAS and A.M PETERSON(2015): "Initial Medication Adherence-Review and Recommendations for Good Practices in Outcomes Research: An ISPOR Medication Adherence and Persistence Special Interest Group Report". Value Health, July, 18(5):690-9.