Use of sequence analysis for classifying individual antidepressant trajectories to monitor population mental health

Analytical methods for identifying sequences of utilization in health data: a scoping review

BACKGROUND

Healthcare, as with other sectors, has undergone progressive digitalization, generating an ever-increasing wealth of data that enables research and the analysis of patient movement. This can help to evaluate treatment processes and outcomes, and in turn improve the quality of care. This scoping review provides an overview of the algorithms and methods that have been used to identify care pathways from healthcare utilization data.

METHOD

This review was conducted according to the methodology of the Joanna Briggs Institute and the Preferred Reporting Items for Systematic Reviews Extension for Scoping Reviews (PRISMA-ScR) Checklist. The PubMed, Web of Science, Scopus, and EconLit databases were searched and studies published in English between 2000 and 2021 considered. The search strategy used keywords divided into three categories: the method of data analysis, the requirement profile for the data, and the intended presentation of results. Criteria for inclusion were that health data were analyzed, the methodology used was described and that the chronology of care events was considered. In a two-stage review process, records were reviewed by two researchers independently for inclusion. Results were synthesized narratively.

RESULTS

The literature search yielded 2,865 entries; 51 studies met the inclusion criteria. Health data from different countries ([Formula: see text]) and of different types of disease ([Formula: see text]) were analyzed with respect to different care events. Applied methods can be divided into those identifying subsequences of care and those describing full care trajectories. Variants of pattern mining or Markov models were mostly used to extract subsequences, with clustering often applied to find care trajectories. Statistical algorithms such as rule mining, probability-based machine learning algorithms or a combination of methods were also applied. Clustering methods were sometimes used for data preparation or result compression. Further characteristics of the included studies are presented.

CONCLUSION

Various data mining methods are already being applied to gain insight from health data. The great heterogeneity of the methods used shows the need for a scoping review. We performed a narrative review and found that clustering methods currently dominate the literature for identifying complete care trajectories, while variants of pattern mining dominate for identifying subsequences of limited length.

Does a working day keep the doctor away? A critical review of the impact of unemployment and job insecurity on health and social care utilisation

While the negative impact of unemployment on health is relatively well established, the extent to which that impact reflects on changes in health and social care utilisation is not well understood. This paper critically reviews the direction, magnitude and drivers of the impact of unemployment and job insecurity on health and social care utilisation across different care settings. We identified 28 relevant studies, which included 79 estimates of association between unemployment/job insecurity and healthcare utilisation. Positive associations dominated mental health services (N = 8 out of 11), but not necessarily primary care (N = 25 out of 43) or hospital care (N = 5 out of 22). We conducted a meta-analysis to summarise the magnitude of the impact and found that unemployed individuals were about 30% more likely to use health services compared to those employed, although this was largely driven by mental health service use. Key driving factors included financial pressure, health insurance, social network, disposable time and depression/anxiety. This review suggests that unemployment is likely to be associated with increased mental health service use, but there is considerable uncertainty around primary and hospital care utilisation. Future work to examine the impact across other settings, including community and social care, and further explore non-health determinants of utilisation is needed. The protocol was registered in PROSPERO (CRD42020177668).

Antipsychotic utilization trajectories three years after initiating or reinitiating treatment of schizophrenia: A state sequence analysis approach

OBJECTIVE

This study aims to describe the utilization patterns of antipsychotic (AP) medication in patients with schizophrenia (SCZ), three years after initiating or reinitiating a given AP.

METHODS

Based on medico-administrative information on patients living in Quebec (Canada), this retrospective cohort study included 6444 patients with a previous diagnosis of SCZ initiating or reinitiating AP medication between January 1, 2012, and December 31, 2014, with continuous coverage by public drug insurance. For each day of follow-up (1092 days), patient was either exposed to one of the chosen categories of APs, or to none. This patient's sequence of AP exposure overtime has been referred to as the "antipsychotic utilization trajectory". These trajectories were analyzed using a State Sequence Analysis, an innovative approach which provides useful visual information on the continuation and discontinuation patterns of use over time.

RESULTS

Clozapine and long-acting injectable second-generation APs had the best continuation and discontinuation patterns over 3 years among all other groups, including less switching of APs, while oral first-generation APs had the poorest patterns. These findings were comparable among incident and non-incident cohorts. Oral second-generation antipsychotics, excluding clozapine, had a poorer continuation and discontinuation pattern than long-acting injectable antipsychotics.

CONCLUSION

State Sequence Analysis provides a clear representation of treatment adherence in comparison with dichotomous indicators of adherence or discontinuation. Consequently, this innovative method has shed light on the impact of the AP chosen to initiate or reinitiate treatment in SCZ, which has been identified as a key factor for long-term treatment continuation and discontinuation.