Quality indicators for penitentiary medical reports supporting the adoption of alternative measures to imprisonment on health grounds

Abstract

Introduction/objectives

Humanitarian release from prison on health grounds is a measure to preserve human dignity, recommended by international human rights bodies and enshrined in the legislation of many countries. A medical report is required for such a judicial decision. The objective of this study was to analyse the quality, content and variability of medical reports used for early release from prison on health grounds.

Methods

A retrospective observational descriptive study was conducted on 164 medical reports for the early release of persons deprived of liberty obtained in the Basque Autonomous Community (BAC) and Catalonia. Structural aspects, clinical content, compliance with the standards of clinical reports in the health system, and relevant medico-legal aspects were evaluated.

Results

The study revealed that there was a high degree of variability, that the importance of the temporary life prognosis for the granting of this type of measure was significant in Spain, and that the health baseline situation of the subjects studied was advanced. Analysis of the quality of the reports revealed significant technical deficiencies and the frequent absence of basic clinical elements. Barriers to accessing information sources limit research.

Conclusions

The absence of standardised guidelines or protocols at the national and international levels for the drafting of this type of report leads us to propose further advancement in this important area of human rights.

Resumen

Introducción/objetivos

La excarcelación humanitaria por motivos de salud es una medida para preservar la dignidad humana, recomendada por organismos internacionales de Derechos Humanos y está recogida en las legislaciones de muchos países. Para dicha decisión judicial se requiere de un informe médico. El objetivo del presente estudio era analizar la calidad, el contenido y la variabilidad de los informes médicos utilizados para la excarcelación anticipada por motivos sanitarios.

Métodos

Se realizó un estudio descriptivo observacional retrospectivo de 164 informes médicos de personas privadas de libertad obtenidos en la Comunidad Autónoma Vasca (CAV) y Catalunya para la excarcelación anticipada por motivos sanitarios. Se evaluaron aspectos estructurales, contenido clínico, la adecuación a los estándares propios de los informes clínicos del sistema sanitario y a aspectos médico-legales relevantes.

Resultados

El estudio reveló que la variabilidad es alta, que el peso que adquiere en España el pronóstico temporal para la adquisición de este tipo de medidas es muy notorio y que la situación basal de los sujetos estudiados es muy avanzada. El análisis de la calidad de los informes evidenció deficiencias técnicas importantes y la frecuente ausencia de elementos clínicos básicos. Los obstáculos para el acceso a las fuentes de información limitan la investigación.

Conclusiones

La ausencia de guías o protocolos estandarizados a nivel nacional e internacional para la redacción de este tipo de informes nos lleva a proponer desarrollar más este ámbito tan importante en materia de Derechos Humanos.

Introduction

The legislation of advanced countries includes, to one degree or another, provisions for release from prison (or the adoption of alternative measures to ordinary imprisonment) on health grounds (Annex A). However, there is no standard term to define this, with terms such as compassionate release, humanitarian release, geriatric release and early release being used, among others.

There is a broad consensus that certain clinical situations, such as deaths in custody, are incompatible with human dignity. This is the view of the Special Rapporteur of the United Nations General Assembly¹ on torture and other cruel, inhuman or degrading treatment or punishment. The standards of the Committee for the Prevention of Torture (CPT),² the European Prison Rules,³ the United Nations Standard Minimum Rules for the Treatment of Prisoners, known as the Mandela Rules,⁴ a resolution of the European Parliament,⁵ and various World Health Organisation-Unites Nations (WHO-UN) publications^6,7 all provide for humanitarian measures such as release from prison on health grounds. Similarly, the European Court of Human Rights (ECtHR) has established that, under specific medical conditions, continued imprisonment may violate human dignity (Article 3 of the ECHR). Furthermore, such conditions can even constitute a violation of the right to life (Article 2) if a direct link exists between inadequate medical care and the prisoner's death.⁸

The three specific elements that the ECtHR takes into account when assessing the compatibility between the state of health and detention are:⁹

• (a)

  the medical condition of the detained person.

• (b)

  the adequacy of medical assistance and care provided during detention.

• (c)

  the appropriateness of maintaining the detention measure in view of their state of health.

In order to establish the right to “adequate” assistance in deprivation of liberty, the CPT proclaimed the “principle of health equivalence”,^4,5,10-12 a principle that has been endorsed by human rights (HR) bodies and the ECtHR itself¹³ and which establishes that healthcare in deprivation of liberty must be equivalent to that received in the outside community.

Despite the legal, ethical and HR relevance of humanitarian release, the data available indicate that it is rarely applied, and the reasons for this have been the subject of international studies. In the United States (US), it is estimated that only 1–4%,^14,15 of applications submitted to the Federal Bureau of Prisons are granted, and rates are similarly low in many state prison systems. Data from European countries, although less accessible, also show that this is an underused resource.¹⁶

A recent US systematic review¹⁵ of trends in humanitarian release policies concludes that it is an underused resource for four main reasons:

• (a)

  Language barrier. The vagueness of regulatory concepts, which require extraordinarily compelling circumstances that are not explicitly stated, and medical terminology, makes them difficult to understand.

• (b)

  Complexity of inclusion criteria, which often require: permanent or irreversible chronic illness, in some cases disability, age, requirement to have served part of the sentence, or economic criteria (availability of family resources for care).

• (c)

  Excessive reliance on the prognosis of the illness.

• (d)

  Social stigmatisation of convicted persons and public apathy regarding their future.

A European study¹⁶ analysing the application of these measures in five legal systems (England-Wales, Spain, France, Germany and Switzerland) concludes that the main barriers to humanitarian release are practical and organisational obstacles (long and complicated process, difficulties in determining definitive prognoses, difficulties in finding care in the community and improving the prison care available), penological goals and multiple interests. It is also hindered by confusion regarding the specific grounds for early release, which are often conflated with justifications such as compassion, the principle of equivalence and other practical concerns like costs and overcrowding.

Regarding the reason for public apathy, Andreas Mitchell and Brie Williams¹⁷ argue that professional medical associations should catalyse and develop mechanisms to promote early release as part of their medical ethics work to protect human dignity. To this end, they propose three lines of action:

• (a)

  Recommend that medical associations lead efforts to promote humanitarian release policies.

• (b)

  Improve palliative care in prisons.

• (c)

  Recommend that those responsible for prison healthcare study and investigate failures and obstacles in order to help develop effective measures and turn them into rules that facilitate procedures.

Although each State establishes its own criteria and requirements for granting humanitarian release, there is one common element: the need for a medical report to justify it. In this regard, it is striking that there is little analysis in scientific literature on the quality or standards that medical reports for release from prison should meet. The above-mentioned study¹⁴ makes several recommendations for their preparation, highlighting six aspects:

• (a)

  Avoid medical terminology: use clear, descriptive, simple and explanatory language.

• (b)

  Accompany the description of the patient's functional status with a descriptive narrative, which is easier to process and remember and is more in line with the principle of humanisation of punishment.

• (c)

  Analyse the prognosis and argue it scientifically.

• (d)

  Enforce the principle of equivalence by identifying community care and assistance standards.

• (e)

  Identify risk factors.

• (f)

  Define extra-community care plans in advance.

• (g)

  Draw up a patient discharge plan.

In the absence of literature, the European Parliament, when questioned on this issue, acknowledges that there is no agreed protocol for the preparation of this type of medical report.¹⁸ In some countries, such as Colombia, forensic institutions have published guidelines for this purpose,¹⁹ recommending that a structure similar to that of a forensic medical report should be followed. And in France, the Ministries of Justice and Health also recommend²⁰ that medical-legal assessments be included in medical reports.

However, unlike other forensic medical reports (for which there are specific frameworks for standardisation and professional development),²¹ the field of medical reporting for humanitarian release does not have its own area of training, research or specialised regulation.

In Spain, only three autonomous communities (AC) or regions (Basque Autonomous Community (BAC), Catalonia and Navarre) have prison health services integrated into the public health system. In the remaining regions, prison healthcare is functionally dependent on the Secretaría General de Instituciones Penitenciarias (SGIP) (Spanish Prison Institution), and in this context, it is the SGIP itself that has published three administrative instructions establishing general criteria for the preparation of medical reports supporting early release on health grounds (Table 1). However, even in the regions where powers have been transferred, there are no guidelines, protocols or standardised structures for their preparation set by respective public health systems departments.

Table 1.

Summary of Spanish prison instructions.

Instruction
01/2000^1	Very serious illness:Estimated risk of death > 10% in 1 year despite treatment.Estimated risk of death > 10% within 5 years despite treatment.Karnofsky index ≤ 50%HIV infection in stage A3, B3 or CChronic psychotic disorder with symptomatic activity despite treatment for > 6 months, or with intellectual impairment
03/2017^2(Repealed)	SeptuagenariansVery seriously ill patients with incurable conditionsImminent danger to life: terminal illness or stage, or death expected with reasonable certainty in the very short term (less than 2 months).^3 With final prognosis report
06/2018^4	SeptuagenariansVery serious illness with incurable conditionsImminent danger to life: very serious and incurable illness with imminent danger to life/terminal condition. With final prognosis report

1

https://www.institucionpenitenciaria.es/documents/20126/78885/c-2000-01.pdf.

2

https://www.acaip.es/images/docs/instruccion_3_2017.pdf.

3

https://www.senado.es/web/expedientdocblobservlet?legis=12&id=38659.

4

https://www.institucionpenitenciaria.es/documents/20126/78885/I_6-2018_Suspensixn_ejecucixn_pena_privativa_libertad.pdf.

Given the importance of humanitarian release in the context of preserving human dignity, its limited practical application and the reasons for this, as well as the central role played by medical reports in its processing, the absence of standardised criteria, specifically the clinical and medical-legal elements that these reports must include, and the absence of studies on their quality, are striking. In this context, the present study aims to describe and assess the quality of reports issued by prison medical staff to request alternative measures to imprisonment for persons deprived of liberty in two of the mentioned AC.

Material and methods

Retrospective descriptive study on the quality of a total of 164 medical reports obtained in the BAC and Catalonia.

Data source

In 2022, we requested medical reports used for humanitarian release procedures from the regional governments of BAC and Catalonia through the Departments of Justice of the Basque and Catalan governments, respectively, which expressed their willingness to collaborate with this work after positively assessing the public and scientific interest of the subject matter. The reports were obtained in 2023. Data that could identify individuals were anonymised by the institutions themselves, which sent the documentation to the authors, who received the reports with the names crossed out. In the case of the BAC, those reports cover the period 2011–2021, while in Catalonia, they correspond to the period 2015–2023.

Only reports from prison doctors were used for the sample; additional reports, such as hospital reports attached in some situations or the single forensic report submitted in one of them, were not taken into account. Similarly, although they are included in the sociodemographic variables, six reports were excluded from the analysis because they lacked a specific prison medical report, providing only clinical records from the hospital where the patient was admitted. This is consistent with the study's objective, which does not focus on the quality of hospital reporting.

The files submitted for the procedure have been analysed, which vary between autonomous communities and between each case. The reports from prisons in the BAC follow the model or form developed by the SGIP (Annex B) and currently in use in the autonomous communities under its jurisdiction. In certain cases in the Basque Autonomous Community, prison doctors added an additional report (prepared by themselves) attached to Annex A. In Catalan prisons, the reports were in free text format, either on a separate sheet or as a continuation of a developmental episode within the medical record.

Variables studied

The following variables were analysed:

• (a)

  Sociodemographic and clinical variables: gender, age, prison, main diagnosis in free text and categorised, and hospitalisation variable (a variable indicating whether the person is hospitalised at the time of the request).

• (b)

  Clinical report quality variables: Created mainly from the sections reflected in Royal Decree 572/2023, of 4 July, establishing the minimum set of data for clinical reports in the Spanish National Health System, the Colombian¹⁹ guide and the French guide.²⁰ The assessment of compliance with the principle of equivalence as a basic ethical principle of medicine has also been incorporated. The presence or absence of these sections and their quality are assessed. An ad hoc scale of qualitative variables has been created (see Table 2).

  Table 2.

  Clinical variables and clinical report quality variables.

  Type of variable	Variable	Description/categories/scale
  Clinics	Main diagnosis	Main clinical diagnosis for which release is requested in free text and grouped into 14 categories. (Cancer,HIV, Hepatitis (B, C), Neurological, Cardiovascular, Respiratory, Multiple pathologies, Other diagnoses, ≥ 70 years old)
  	Hospitalisation	Yes/No. If the patient was hospitalised at the time of the request for release from prison
  Report quality	Background/other diagnoses	Ad hoc scale: Scale created based on whether the following items are present and their corresponding score – Family history (0.1), – Social and professional history (0.1), – Previous illnesses and medications (0.2) – Neonatal, obstetric, and surgical history (0.2) – Allergies (0.1) – Toxic habits (0.2) – Preventive measures (0.1)
  	Prognosis	Yes/no: Presence or absence in the patient's prognosis. Temporality (short, medium, long); If the temporary prognosis is describedDescriptive prognosis: If the forecast is developed and argued – No description – Only a brief descriptive mention – Detailed explanation Prognosis scale: yes/no If a prognostic scale has been used
  	Current history and evolution	According to the scale created: – Summary of current episode: If a summary of the current episode has been made o Does not contain o Mention o Developed – Progress and comments: If a summary of the progression and comments on the disease has been made o None o Mention o Detailed explanation
  	Physical examination	Yes/No Presence or absence of a description of the physical examinationGeneral examination Yes/NoExamination by affected organ Yes/No
  	Function	Present/Absent Presence or absence of a description of the patient's functional statusScales yes/no and scale mean – Karnofsky – Barthel), – Other: other scales or any mention or description of functionality.
  	Additional tests	Yes/No If the additional tests performed are mentioned
  	Treatment	Yes/No If the treatments the patient is taking are mentioned
  	Additional report	Yes/No; If any additional reports are addedSource: – own compilation – public administration – both
  	Medical-legal recommendations	Yes/No: whether it contains any medical-legal considerations or recommendations
  	Principle of equivalence	Yes/No: if any assessment is made of the principle of equivalence (regarding healthcare received in prison)

Ethical aspects

The study was submitted to the Ethics Committee (M10 2022 127) of the University of the Basque Country and was approved on 3 June 2022. All data have been anonymised, and the database generated complies with the requirements of the Organic Law on Data Protection.

Statistical analysis

A descriptive analysis of all variables was performed, expressing quantitative variables as the arithmetic mean and standard deviation, and categorical variables as frequencies and percentages. The SPSS statistical package, version 20.0, was used to process the data. Differences in the characteristics of the clinical reports between Catalonia and the BAC were compared using Fisher's exact test. A p-value of < 0.05 was considered statistically significant.

Results

The sample analysed consisted of 164 medical reports from the BAC (n = 73) and Catalonia (n = 91). Fig. 1 shows the flowchart of the reports that make up the final sample.

Figure 1.

Flowchart of the reports.

Table 3 shows the descriptive analysis of the sample under study, the majority of which were male (80.5%), while only 9.1% were female. In 10.4% of cases, the sex of the ill person was not recorded or could not be deduced from the wording of the report. The average age was 54.3 (sd 12.8) years.

Table 3.

Description and results of sociodemographic and clinical characteristics of the sample.

Variable	N
Age (1)	54.3 (sd12.8)
Gender
Male	132 (80.5%)
Female	15 (9.1%)
Not specified in report	17 (10.4%)
Prison
BAC	42 (49.9%)
Catalonia	75 (45.3%)
Other regions	8 (4.8%)
Main diagnosis by pathological group
Cancer	49 (29.9%)
HIV	7 (4.3%)
Hepatitis (B, C)	3 (1.8%)
Neurological	16 (9.8%)
Cardiovascular	16 (9.8%)
Respiratory	7 (4.3%)
Multiple pathologies	61 (37.2%)
Other diagnoses	3 (1.8%)
≥ 70 years old	1 (0.6%)
Hospitalised
Yes	29 (17.7%)
Not explicitly stated, but inferred (no hospitalised) from the documentation	22 (13.4%)
No	80 (48.8%)
Unknown	33

(1) In the case of age, the mean and standard deviation are shown.

(2) Prisons in Catalonia (Brians 1, Brians 2, Wad Ras, Lledoners, Más d, Enric Tarragona, Ponent de Lleida, Puig de las Bases de Figueres, Quatre Camins, La Model). Prisons in the Basque Autonomous Community (Zaballa, Martutene, Basauri). Other communities (Brieva, Dueñas, El Dueso, León, Pamplona), files of persons residing in the Basque Autonomous Community who, at the time of the report, were admitted to these prisons.

(3) Other diagnoses (digestive, endocrine-metabolic, septuagenarians).

With regard to the distribution by prison, 45.3% of the reports came from Catalonia and 47.5% from the BAC. A total of 4.8% were reports made in other AC corresponding to patients from the BAC who, at the time the report was issued, were admitted to prisons in other communities and whose files were subsequently transferred.

Regarding the hospital situation at the time of the request for release, 62.2% were not hospitalised (including cases where this was explicitly stated, as well as those where it can be inferred from the documentation). Also, 17.7% were hospitalised at the time of the request, and in 20.1% of cases, this information was unknown.

As for the main diagnosis, the most common were multiple pathologies (37.2%), followed by cancer (29.9%). Among the cancer diagnoses (n = 49), it is also worth noting that most were in metastatic stages (51%) when the report was issued (14 lung, 1 breast, 1 Gastrointestinal stromal tumour (GIST), 2 renal, 1 colon, 1 prostate, 1 melanoma, 1 urothelial, 1 hepatocarcinoma, 1 without primary site). The rest were histologically aggressive or in advanced stages (locoregional involvement). Of the fifteen reports on women, two had a diagnosis of cervical cancer. Neurological and cardiovascular diseases each accounted for 9.8% of cases. Other categories, such as HIV and hepatitis, although present in patients with multiple pathologies, were less frequent as a primary diagnosis.

Table 4 describes the main characteristics of the clinical reports. A particularly relevant finding is that none of the 164 reports contained any description or comment on the principle of equivalence.

Table 4.

Description of the quality of the clinical reports in the sample.

Variable	n (%) X¯ (sd)
Personal history
Present	122 (77.2%)
Absent	36 (22.8%)
Quality of personal background	4.2 (sd 2.1)
Current illness, summary of episode and evolution
Present	110 (69.6%)
Absent	48 (30.4%)
Episodic summary available
None	4 (3.7%)
Present	36 (33.3%)
Developed	68 (62.9%)
Progression, complications, comments
None	16 (14.9%)
Present	21 (19.6%)
Developed	70 (65.4%)
Physical examination
General examination
Present	11 (7.0%)
Absent	147 (93.0%)
Affective apparatus	5 (3.2%)
Psychopathological examination	2 (1.3%)
Prognosis
Present	120 (75.9%)
Absent	38 (24.1%)
Temporary description prognosis
Present	109 (68.4%)
Short	67 (61.4%)
Medium	39 (35.7%)
Long	3 (2.8%)
Absent	49 (31%)
Description
Present	61 (38.8%)
Detailed explanation	16 (26.3%)
Mention	45 (73.8%)
Absent	97 (61.3%)
Prognostic scale
Present	2 (1.3%)
Absent	156 (98.7%)
Summary complementary tests
Present	31 (19.6%)
Absent	127 (80.4%)
Treatments
Present	62 (39.2%)
Absent	96 (60.8%)
Recommendations and medical-legal considerations
Present	82 (51.9%)
Absent	75 (47.5%)
Functionality
Present	104 (66.4%)
Absent	53 (33.5%)
Scale used	n(%)/mean on scale
Karnofsky index	75 (71.4%)/51.2 (sd 19.7)
Barthel Index	6 (5.7%)/45.0 (sd 20)
None	23 (22.1%)
Additional report
Present	91 (57.6%)
Absent	67 (42.4%)
Type of report
Prepared in-house	44 (48.3%)
Public Health	34 (37.4%)
Both	13 (14.3%)
Absent	67 (42.4%)

While most reports lack a dedicated section for medical history, 77.2% include relevant information on medical history or other diagnoses; however, the average quality of information was rated 4.2 out of 10 according to the assessment scale used.

In relation to the description of the current illness, 69.6% of the reports contained some mention of the current clinical episode, while 30.4% did not include information on this regard.

It was observed that the general physical examination was only described in 7.0% of the reports, and the section of complementary tests performed was absent in 80.4% of cases.

The mention of the treatment section in reports was included in the 39.2% of the cases. As regards regions, it was included in the 47.9% of reports in Catalonia and 31.7% in the BAC, and this regional difference was found to be statistically significant (p = 0.025).   

The 51.9% of reports included a comment or sentence on a medical-legal recommendation or consideration, although most did not take an explicit position.

The prognosis section was one of the most developed: it was included in 75.9% of reports. There were notable differences between regions: 90.6% of reports from the BAC included a prognosis, compared to 58.9% of those from Catalonia (p < 0.001). Of the reports that address prognosis, 90.0% refer to life expectancy, but only 1.6% based this on a validated prognostic scale. In terms of content, 36.6% provided a brief explanatory mention of the prognosis. Of the 109 reports containing a time prognosis, 61.4% had a poor short-term prognosis, 31.7% a poor medium-term prognosis and 2.8% a poor long-term prognosis.

Patient functionality was recorded in 104 patients (66.4%), with 71.4% using the Karnofsky index and 5.7% using the Barthel scale. A total of 22.8% provided a brief description of the patient's functionality without using any scale. Also worth mentioning were the average functionality scales, 51.2 on the Karnofsky index and 45 on the Barthel index.

Finally, 52.6% of the reports included an additional report, with clear regional differences: 81.9% of those in the BAC compared to 19.2% in Catalonia (p = < 0.001).

Discussion

To the best of our knowledge, our study is the first of its kind, which reviews the overall quality of medical reports for humanitarian release. We believe it provides a future line of work to improve these documents in the prison health care setting.

Compared to the limited literature available, our results are consistent with those published by Handtke et al. and Kaushik et al.,^16,15 as the reports show an excessive reliance on prognosis. There is also significant morbidity and mortality in this population, as demonstrated by numerous studies and statistics.²²

Although the provisions of the criminal code are generic, the public health departments of Catalunia and BAC have not established guidelines, and the SGIP's prison instructions place special emphasis on prognosis, our study shows that prognosis takes on great importance in the implementation of humanitarian release, possibly to the detriment of other factors that could be relevant to making this type of complex medical decision (quality of life, principle of equivalence, relevant prison record, etc.). This limits the comprehensive perspective needed for an adequate assessment, which could result in this procedure being reduced to a mere formality to prevent deaths in prison.

This is particularly evident in some cancer cases, as the non-metastatic cases in the sample were mostly cancers with an initially poor survival prognosis, and it can be inferred from the data provided that, prior to the diagnosis of metastasis leading to the request for release, the individuals concerned had received cancer treatment while in prison (surgery, chemotherapy and radiotherapy). However, this has not been confirmed in all cases due to a lack of information in the reports.

The high incidence of cervical cancer in this sample is also striking (both stage IIB: estimated survival rate 58–63.9%,^23,24 considering that the population incidence is 5.2 cases per 100,000 women,²⁵ that there is population screening, and that both false negatives and interval cancers are very rare (less than 0.5%²⁶). However, due to incomplete data in the reports, it is not possible to discern whether these results stem from disparities in the access to or quality of prison healthcare, or from alternative causes (e.g., patient refusal of cytological screening, false negatives, or interval cancers).

The study shows considerable room for improvement in the quality of reports. The frequent absence of basic sections of a medical report, such as physical examination, complementary tests, current treatments, and deficiencies in the current episode section and the poor quality of the medical history and other diagnoses, deserve critical reflection and are a clear area for improvement. Given that the recipients of the conclusions of this type of report are non-healthcare professionals—a circumstance that requires clear and accessible writing of the conclusions, avoiding excessive use of technical terms—this aspect has predictably taken on disproportionate importance, to the detriment of more technical sections (such as those mentioned above), which are precisely those that provide the conclusions with the necessary support, objectivity and traceability.

There are also basic deficiencies in demographic data, such as the sex of the person or whether the patient is hopsitalised or not. The absence of an assessment of the principle of equivalence is worrying, given its ethical and legal relevance in the prison context.

Despite the fact that the principle of integration^5,11,27,28 in public health is considered the most effective measure for promoting the principle of health equivalence in prisons, and although Law 16/2003 on the cohesion and quality of the National Health System provided for the transfer of powers to the AC in the area of prison health in general (and with it the integration of prison health services into the National Health System). In Spain, this transfer has only been carried out in three AC, namely Catalonia, Navarre and the BAC. Anonymised prison medical reports were also requested from the General Secretariat of Penitentiary Institutions. This request was denied on technical grounds, as it was not feasible to extract the requested data in a standardised manner. Our main limitations lie in the small size of the sample, as a result of the bureaucratic and administrative obstacles we encountered in accessing sources of information, due in part to the absence of data digitalization procedures. The fact that we only have reports from two of the 17 AC that make up the State, and that these are the ones that comply with the principle of healthcare integration, limits the generalisation of our findings.

Another relevant limitation is that we have had to generate our own ad hoc evaluation criteria or rely on existing evaluation criteria common to medical records, as we did not have standardised reports agreed upon by scientific societies on the standards that such reports should meet.

The statistically significant differences (in the prognosis and treatment sections) found in the bivariate analysis between Catalonia and the BAC may be due to the format used (the automatic transfer of treatment to the reports in Catalonia and to the specific prognosis section in the forms in the BAC), and it could be inferred that having a concise and undeveloped model or form greatly conditions a report, and that free text can lead to a great deal of variability, allowing the development of certain items, as well as the absence of other relevant ones. With the development of a guide or protocol that unifies criteria and facilitates comparability, an interchangeable, verifiable and more homogeneous document would be achieved, which would allow for research and improvement of the measure of humanitarian release.

The absence of standard terminology and coding to refer to the adoption of alternative measures to closed prison regimes for health reasons may be contributing to their low visibility and application, increasing social and medical stigma and apathy towards a practice that is essential for ensuring humane and dignified care for these ill individuals in prison. The absence of uniform and appropriate language that gives it relevance in the academic sphere also hinders its research and development.

Overall, these findings highlight the urgent need to improve the quality, standardisation and transparency of prison medical reports in order to optimise humanitarian release and guarantee the rights and health of persons deprived of liberty.

Conclusions

Limited access to medical reports and the absence of standardised protocols are significant barriers to rigorous research and the development of best practices in medical assessment for humanitarian release. The reports evaluated show low overall quality and are overly reliant on life expectancy, which may restrict a comprehensive assessment. The implementation of uniform criteria and standardisation of report formats could contribute to better application of early release and the humanisation of prison sentences.

Recommendations and future action

Based on our findings, the following is recommended:

• 1.

  Standardisation and protocolisation: Develop and promote the adoption of a single, protocolised model for the preparation of clinical reports for humanitarian release, which comprehensively considers clinical, social and legal criteria.

• 2.

  Specialised training: Train the professionals involved (prison doctors, forensic doctors and clinicians) in the correct preparation and assessment of these reports, emphasising multidimensional aspects of patient health and well-being.

• 3.

  Improved accessibility and recording: In order to impove future analysis, it would be advisable to implement specific coding and archiving systems for these reports in the medical databases of all entities involved in the humanitarian release procedure, facilitating access for research and monitoring.

• 4.

  Continuous research: Promote multicentre and collaborative studies to expand knowledge about the quality and effectiveness of these reports in different autonomous communities and prison contexts.

• 5.

  Incorporation of ethical and human rights criteria: Ensure that reports include a comprehensive assessment that considers not only the medical prognosis but also the quality of life and dignity of the person, in line with the principles of equivalence and respect for fundamental rights.