Original Article

Socio-demographics characteristics and health conditions of older homeless persons of Lima, Peru

Marlene Moquillaza-Risco^1,a, Elsa Leon ^1,a, Mario Dongo ^1,b, Cesar V. Munayco ^1,c

¹ Programa Nacional Vida Digna (PNVD), Ministry of Women and Vulnerable Populations. Lima, Peru.

^a Licentiate in Social work; ^b environmental engineering undergraduate student; ^c physician, master’s degree in epidemiology and public health

ABSTRACT

Objectives. Determine the socio-demographics characteristics and health conditions of older homeless persons at the time of enrollment into the National Program “Vida Digna” and the probability of functional dependency by age, and stratified by gender and cognitive impairment. Materials ande methods. We performed a cross sectional study, reviewing all registration forms of the program in order to identify socio-demographic variables and health conditions of older homeless persons at the time of enrollment in the program. We did a descriptive analysis of the socio-demographic variables and we also determined the frequency of health conditions. Furthermore, we determined the probability of functional dependency by age, and stratified by gender and cognitive impairment through a logistic regression model. Results. The older homeless persons at the time of enrollment in the program were mostly single men, with a primary education or no education. The study subjects had a high frequency of chronic and mental diseases. 50% of them had certain level of functional impairment and roughly 70% had a certain level of cognitive impairment. The probability of functional dependency increased by age, and it was higher in women than in men. This probability increased according to the level of cognitive impairment. Conclusions. This study shows that older homeless persons are a vulnerable population not only because they live outdoors but also because they a have also for the high prevalence of chronic and mental diseases. These diseases prevent the homeless persons from living by themselves special care to overcome their situations.

Key words: Elderly; Homeless person; Aging; Mild cognitive impairment (source: MeSH NLM).

INTRODUCTION

Aging is a natural and irreversible physiological process. It is a triumph of modern healthcare policies that people live longer lives ⁽¹⁾. This situation results in new challenges for the Peruvian government, who must provide the necessary services for the protection and care of senior citizens so that they may live longer with good quality of life. In this sense, the government must create public policies guaranteeing that older people lead pleasant, active, and independent lives within their families and communities ^(2,3).

Unfortunately, there is a group of senior citizens who are invisible to society, live in highly hazardous social situations, and are constantly at risk, who have been excluded and disconnected from their families, and are often not respected in their social environment ⁽²⁾. This group are homeless senior citizens (PAMSCs, Spanish acronym) who not only live in extreme poverty but under nearly inhumane conditions ^(4-6), without a home or basic utilities, no satisfaction of basic needs, no access to health care or recreation, abandoned and forgotten by their families and the government ^(7-9).

Homeless people, aside from the normal deterioration that comes with age, are also at an unacceptably high risk of preventable diseases (pneumonia, influenza), addictions (drug addiction, alcoholism), victimization (sexual assault), progressive morbidity (e.g., chronic diseases, cancer), and life expectancy below the national average ^(4,10).

Currently, there is solid evidence for the success of social programs for homeless individuals ^(11-13). This evidence indicates that shelter and adequate management of homeless people lead to social reintegration, reduced numbers of visits to emergency rooms and hospitalizations due to fewer complications, addiction control, and above all, increased life expectancy and improved quality of life ^(14-20).

One of the decisive factors for the social/familial reintegration of homeless people, especially senior citizens, is their degree of functional and cognitive independence, that is, their ability to execute basic activities of daily living (ADL) such as getting dressed, eating, or bathing; this parameter encompasses the concept of instrumental ADL (IADL), such as shopping or preparing food ⁽²¹⁾.

In 2012, the Peruvian government, in an effort to address this problem, created the National Decent Life Program (Programa Nacional de Vida Digna; PNVD) in order to restore and protect the rights of PAMSCs. The PNVD offers restoration of their right to identity, access to the Comprehensive Healthcare System (SIS, Spanish acronym), permanent or nocturnal shelter, food, clothing, and age-appropriate care ⁽²²⁾. The PNVD currently serves 457 PAMSCs in the cities of Lima, Arequipa, Huaraz, and Huanuco.

In Peru, despite the government’s efforts to address this problem, little is known about the above-mentioned group of people because they are a mobile segment of the population and hard to approach. For this reason, with the goal of obtaining evidence to help implement better public policies for preventing and overcoming this problem, the objective of this study was to determine the sociodemographic and basal health characteristics of PAMSCs by analyzing PNVD beneficiaries at the time of admission into the said program. In addition, we determined the probability of functional dependence due to age and stratified the data by gender and cognitive impairment.

MATERIALS AND METHODS

STUDY DESIGN

We conducted a cross-sectional study, where we reviewed PNVD archives to find the following sources of data: initial contact sheets, an economic assessment sheet, health sheet, and social and psychological reports for each PNVD beneficiary.

STUDY AREA AND POPULATION

This study was conducted on homeless adults over 60 years old who entered the PNVD between 2012 and June 2014 at all locations of the program (Lima, Peru) and who were housed at some Residential Care Center for Senior Citizens (CARPAM, Spanish acronym) or some guest house at the national level. CARPAMs offer permanent accommodations, nursing care, clothing, full meals, and health care. By contrast, guest houses offer overnight accommodations, clothing, food, and health care. Both services also offer workshops and recreational activities to strengthen abilities that optimize the aging process.

DATA COLLECTION

Between June and July 2014, PNVD archives were analyzed to identify the sources of information and to determine the baseline of the program in terms of the characteristics of the beneficiaries. We identified five sources of information, which are described below.

The initial contact sheet is filled out by PNVD workers who are in charge of approaching homeless senior citizens in order to encourage them to leave the streets and invite them to the PNVD. This sheet contains demographic information (age, gender, educational level, marital status, and place of birth), whether they have a national ID, where they sleep, whether they have relatives and some type of insurance, or whether they have some disease. This sheet serves as a screening procedure to identify potential PNVD beneficiaries. It also collects self-reported information about the beneficiary’s health.

The economic assessment sheet evaluates the type of job they have and whether they are affiliated with some sort of social program. The health sheet makes an overall assessment of the beneficiary’s health and dependence/independence status through KATZ’s index of independence. This test evaluates the degrees of dependence or independence using six basic functions: bathing (a sponge, shower, or tub), getting dressed, using the toilet, mobility, continence, and eating ⁽²³⁾. In this test, each activity is classified into three categories: independence, partial dependence, and complete dependence.

According to KATZ’s index, a person is considered independent when he or she does not need help or needs help only for one component of the activity; someone is considered completely dependent when they need someone else’s help, including supervision or guidance, for all components of the activity. Finally, someone is partially dependent when they “require assistance” for more than one component but can carry out other components of the activity without help or supervision ⁽²⁴⁾.

The social evaluation determines whether the person indeed meets the established PAMSC profile, whether they are socially excluded (essential requirement to enter the PNVD); and the psychological sheet evaluates, among other parameters, cognitive status by Pfeiffer’s test, and the emotional status according to the Yesavage Geriatric Depression Scale. Pfeiffer’s test evaluates cognitive impairment by means of a 10-item questionnaire on a small number of relatively basic functions (short- and long-term memory, attention, orientation, information about daily events, or mathematics proficiency) ⁽²⁵⁾.

This scale categorizes the cognitive level as follows: normal cognition, mild cognitive impairment, moderate (pathological) cognitive impairment, and severe cognitive impairment.

Yesavage’s scale is applied to the elderly and is focused on assessing the affective state, that is, geriatric depression ⁽²⁶⁾. It is important to keep in mind that the results of these tests were collected from the aforementioned sheets.

DATA ANALYSIS

The information from the different sheets was entered into a Microsoft Excel database and converted to CSV format for analysis by means of the epicalc package of the R programming language ⁽²⁷⁾. To protect personal information of the PNVD beneficiaries, names and last names were encrypted and were assigned an identification code.

We performed a descriptive analysis based on the frequency distribution of the main sociodemographic variables, such as gender, education level, and marital status. We also calculated the median and the minimal and maximal values for the age variable. We determined the relative frequency of physical and mental disorders in the subjects as well as their dependence status and mental state.

We determined the probability of partial dependence (there were no completely dependent beneficiaries) by age and by gender, by means of a logistical regression model. Finally, we determined the probability of partial dependence by age and degree of cognitive impairment.

ETHICAL CONSIDERATIONS

This study’s protocol was approved by the Executive Directorate of the National Decent Life Program (permission to review their archives). To ensure confidentiality of the beneficiaries’ personal information, we did not record any information that could identify them and instead used identification codes.

RESULTS

We collected information from 348 PNVD beneficiaries, and 46 were excluded for not knowing or being uncertain about their age, or if they were under 60 years old (one of the criteria for entering the PNVD is being older than 60) or over 100 years old (a probable error in the record).

The median of age of the beneficiaries at the time of admission into the PNVD was 72 years, and most (82.8%) were males. Regarding the education level, 37.8% had completed or attended primary school, and 10.9% attended no classes. It should be pointed out that 8% had college education. Most of the beneficiaries were single (69.6%), whereas only 15.8% were married. There are many ways in which PAMSCs are identified and referred to the PNVD. In our case, most PAMSCs were identified or referred by program facilitators (65.6%) or by other institutions such as police (7.4%). Only 5% came of their own accord (Table 1).

Until June 2014, 47.7% of the PNVD beneficiaries were staying at a CARPAM or using the guest house service (25.7%); 19% and 7% had left the PNVD or died, respectively (Table 1).

Regarding the family network of the PAMSCs, only 29 (9.6%) of them had some kind of a close relative. Among their close relatives, 15 (5.0%) were children, 9 (3.0%) were siblings, and 3 (1.0%) were nephews or nieces (data not included in Table 1).

We found that 66.2% of the PAMSCs had some chronic disease at the time of admission into the PNVD, e.g., hypertension (14.2%), arthritis/osteoarthritis (7.6%), or diabetes mellitus (5.6%). Some of them also had a history of infectious diseases, such as syphilis (5%) or tuberculosis (2.9%; Table 2). It should be noted that many of the PAMSCs had several diseases simultaneously. In addition, some of the PAMSCs had problems with alcohol or drugs such as marijuana or cocaine paste. Furthermore, 48.9% of the PAMSCs were partially dependent upon admission into the PNVD.

Mental disorders were also prevalent in this group of vulnerable people: 27.8% of them had some type of mental or neurological disorder at the time of admission into the PNVD, mainly senile dementia (10.3%), some type of schizophrenia (6%), or Parkinson’s disease (3%). Finally, 23.2% and 12.5% of the PAMSCs showed moderate and severe cognitive impairment, respectively, at the time of admission into the PNVD (Table 2).

The social and family assessment of the PAMSCs revealed that all of them were at risk of social problems because of the absence of support by their families or a social support network. Furthermore, Yesavage’s test showed that all of the PAMSCs were depressed to some extent (data not shown in Table 2).

The probability of partial dependence increased with age, both among men and among women. Additionally, women were found to be more likely than men to become partially dependent (Figure 1). For example, the likelihood of partial dependence increased from 0.75 at 60 years of age to nearly 0.90 at 80 years of age. By contrast, the likelihood of partial dependence in men increased from 0.6 at 60 years of age to nearly 0.70 at 80 years of age.

The likelihood of partial dependence increased with age for all degrees of cognitive impairment, except for severe cognitive impairment, where the likelihood of partial dependence was close to 1.0 at all ages. On the other hand, partial dependence increased with the severity of cognitive impairment (Figure 2). For example, in PAMSCs without cognitive impairment, the likelihood varied between 0.4 and 0.5 between 60 and 80 years of age, whereas among PAMSCs with cognitive impairment, this likelihood varied between 0.6 and 0.7 between 60 and 80 years of age. Among PAMSCs with moderate cognitive impairment the likelihood varied between 0.8 and 0.95 within this age range.

DISCUSSION

Thus study shows that PAMSCs admitted to the PNVD are mostly are male, single, and have an education level of primary school or below. These PAMSCs show high prevalence of chronic diseases, some of the most salient ones being hypertension, arthritis or osteoarthritis, and diabetes mellitus. They also show high prevalence of mental and neurological disorders such as senile dementia, schizophrenia, and Parkinson’s disease.

Furthermore, the results of this study show that nearly 50% of the PAMSCs admitted to the PNVD are partially dependent, and nearly 70% show some degree of cognitive impairment. Finally, the probability of partial functional dependence increases with age, and we found that this probability is higher among women than among men; in addition, this probability increases greatly with the severity of cognitive impairment.

This study has the following limitations: 1) Because the data were collected from the PNVD registration sheets, our results may underestimate the actual data (not accurate results), mainly in relation to the prevalence of chronic diseases and mental/neurological disorders because this information does not come from an exhaustive medical examination but rather from the report of the PAMSC or from an assessment by the PNVD employees who are not healthcare professionals. Despite this problem, there are diseases (such as the ones we mentioned above) that are easily recognizable by both the patient and the PNVD employees, especially in senior citizens. 2) Another significant limitation is that persons with cognitive impairment and mental disorders cannot provide accurate data. 3) The information about age may be inaccurate, mainly among persons who cannot provide this information or refuse to do so. It is difficult to estimate age in this group of people during a general physical examination because many of them look older than they are due to the bad living conditions. In many cases, the real age is identified after the PAMSC has entered the PNVD, and some trust has been established between the PAMSC and the PNVD team, or when the information is verified by the National Register of Identification and Civil Status (RENIEC, Spanish acronym). This is also true for the data on relatives.

A common technology for obtaining information and for identification of homeless people is a census or registry, which are population studies focused on a specific geographic area, such as a city. This was the case for the census in Medellin, Colombia, in Mexico City, and in El Alto in Bolivia; these projects can even encompass an entire country, such as the registries compiled in Chile in 2005 and 2011. These studies showed the same demographic pattern predominance of single males) and educational pattern (primary education or less) for PAMSCs as in our tudy ^(7-9,28). This pattern may be explained as follows: single males have fewer familial connections, and the low educational level prevents them from finding well-paid jobs that can provide for their basic needs.

This study shows the importance of PNVD promoters for identification and enrolling of PAMSCs as well as the usefulness of other governmental and civil-society agencies because PAMSCs are difficult to identify and approach due to their high mobility and evasion of authorities; in many cases, a census or registry can help to identify PAMSCs ^(29,30).

Undoubtedly, this particular group of senior citizens shows higher prevalence of chronic diseases, mental disorders, infectious diseases, and alcohol and drug abuse than does the general population of senior citizens ^(2,4,31). Unfortunately, due to the limitations of self­reporting, in some cases, these prevalence values are below what is expected for senior citizens, as is the case for hypertension ⁽³¹⁾.

Censuses and registries in our continent do not include health information, which is why we cannot compare our results to other studies continent-wide; therefore, we will compare our results to those of the Health, Welfare and Aging Survey (SABE, Spanish acronym) conducted in several Latin American and Caribbean countries ⁽³¹⁾ among the elderly. For example, the SABE study indicates that, on average, the prevalence of chronic diseases varies between 10% and 40% among 60- to 80-year-olds, whereas in our study, it was 66.2%. That study also indicated that prevalent diseases at this life stage are hypertension, cardiovascular disease, arthritis, and diabetes, in line with the results of our study. Here, mental disorders were found to be more prevalent than in the SABE study ⁽³¹⁾; overall, the prevalence of mental disorders among PAMSCs at the time of admission into the PNVD is ~30%. In contrast, according to the SABE study, the prevalence of depression varies between 5% and 20% (in our study it is 100%). Our findings differ from those of studies conducted on PAMSCs in other zones: in the United States, Australia, and Europe, where the prevalence of depression varies between 50% and 60%, and the prevalence of mental disorders is 61% ^(4,32,33).

The authors of the SABE study also assessed functional dependence and found that between 10% and 25% of senior citizens have at least some kind of difficulty with ADLs and IADLs ⁽³¹⁾. Here, we found that 50% of these people have partial functional impairments. In addition, the frequency of cognitive impairment is fourfold higher among PAMSCs than among the senior citizens surveyed by the SABE. Our results match those of other studies on the same homeless population ⁽³⁴⁾.

Another important result of this study is the finding that partial functional impairment is directly related to age and the degree of cognitive impairment, in agreement with research on aging ^(11,35). Our study also shows that women are much more dependent than men because women show a higher probability of functional impairment, probably because they have higher prevalence of arthritis, osteoarthritis, and osteoporosis among other illnesses ⁽³¹⁾.

This study, despite the limitations due to its cross­sectional design and the review of registration sheets, provides information that is valuable not only for the PNVD but also for the Peruvian government and society at large because our results represent the first sociodemographic description and identify the physical- and mental-health problems among PAMSCs in Peru. Fortunately, these problems are being addressed by the government through the PNVD. We also demonstrated that functional impairment and cognitive impairment are useful metrics for the PNVD because the care and services provided to this vulnerable group of people depend on these parameters, as does the distribution of the budget that helps to fulfill the relevant needs.

Although this study provides data on PAMSCs, it is necessary to conduct further research to elucidate the reasons why these Peruvians end up living on the streets. It is also necessary to organize initiatives for evaluation of the cost-effectiveness of social programs for vulnerable populations such as homeless senior citizens.

The results of this study may help to increase the continuity and scale and to enhance advocacy for social programs whose objective is to improve the rights and living conditions of this highly vulnerable group of Peruvians.

Author Contributions: MMR and CVM contributed to the conception of the study and writing of the manuscript. MMR obtained the funding. EL and MD collected the data and reviewed the manuscript. CVM performed the data analysis.

Sources of Funding: this study was funded by the PNVD of the Ministry for Women and Vulnerable Populations (MIMP, Spanish acronym).

Disclaimers: the contents of this manuscript are the sole responsibility of its authors and do not necessarily represent the official opinion of the PNVD of the MIMP.

Conflicts of Interest: the authors declare that they do not have any conflicts of interest.

References

1. Kelly PJ. Aging -- a public health success . Public Health Nurs. 2014 Jul-Aug;31(4):289. doi: 10.1111/phn.12132.

2. Cohen CI, Sokolovsky J, Crane M. Aging, homelessness, and the law . Int J Law Psychiatry. 2001 Mar-Jun;24(2-3):167-81.

3. Satariano WA, Guralnik JM, Jackson RJ, Marottoli RA, Phelan EA, Prohaska TR. Mobility and aging: new directions for public health action . Am J Public Health. 2012 Aug;102(8):1508-15. doi: 10.2105/AJPH.2011.300631.

4. Fischer PJ, Breakey WR. The epidemiology of alcohol, drug, and mental disorders among homeless persons . Am Psychol. 1991 Nov;46(11):1115-28.

5. Freeman RB, Hall B. Permanent homelessness in America?. Population Research and Policy Review. 1987;6(1):3-27.

6. Kondratas A. Ending homelessness. Policy challenges . Am Psychol. 1991 Nov;46(11):1226-31.

7. Instituto de Asistencia e Integración Social. Censo "Tú también cuentas IV". 2011-2012. México: Instituto de Asisitencia e Integración Social; 2012.

8. Ministerio de Desarrollo Social. En Chile todos contamos. Segundo catastro nacional de personas en situación de calle . Chile: Ministerio de Desarrollo Social; 2011.

9. Ministerio de Planificación. Habitando la calle. Catastro Nacional de Personas en Situación de Calle, 2005 . Chile: Ministerio de Planificación; 2005.

10. Centers for Disease Control (CDC). Deaths Among Homeless Persons -- San Francisco, 1985-1990 . MMWR Morb Mortal Wkly Rep. 1991 Dec 20;40(50):877-80.

11. Plumb JD. Homelessness: care, prevention, and public policy . Ann Intern Med. 1997 Jun 15;126(12):973-5.

12. Daiski I. Perspectives of homeless people on their health and health needs priorities . J Adv Nurs. 2007 May;58(3):273-81.

13. Dickey B. Review of programs for persons who are homeless and mentally ill . Harv Rev Psychiatry. 2000 Nov;8(5):242-50.

14. Tsai J, Mares AS, Rosenheck RA. Does housing chronically homeless adults lead to social integration?. Psychiatr Serv. 2012;63(5):427-34. doi: 10.1176/appi.ps.201100047.

15. de Vet R, van Luijtelaar MJ, Brilleslijper-Kater SN, Vanderplasschen W, Beijersbergen MD, Wolf JR. Effectiveness of case management for homeless persons: a systematic review . Am J Public Health. 2013 Oct;103(10):e13-26. doi: 10.2105/AJPH.2013.301491.

16. Rodriguez RM, Fortman J, Chee C, Ng V, Poon D. Food, shelter and safety needs motivating homeless persons’ visits to an urban emergency department . Ann Emerg Med. 2009 May;53(5):598-602. doi: 10.1016/j.annemergmed.2008.07.046.

17. Burt MR. Impact of housing and work supports on outcomes for chronically homeless adults with mental illness: LA’s HOPE . Psychiatr Serv. 2012 Mar;63(3):209-15. doi: 10.1176/appi.ps.201100100.

18. Martinez TE, Burt MR. Impact of permanent supportive housing on the use of acute care health services by homeless adults . Psychiatr Serv. 2006 Jul;57(7):992-9.

19. Clark C, Rich AR. Outcomes of homeless adults with mental illness in a housing program and in case management only . Psychiatric Serv. 2003 Jan;54(1):78-83.

20. Kessell ER, Bhatia R, Bamberger JD, Kushel MB. Public health care utilization in a cohort of homeless adult applicants to a supportive housing program . J Urban Health. 2006 Sep;83(5):860-73.

21. Dorantes-Mendoza G, Ávila-Funes JA, Mejía-Arango S, Gutiérrez-Robledo LM. Factores asociados con la dependencia funcional en los adultos mayores: un analisis secundario del Estudio Nacional sobre Salud y Envejecimiento en Mexico, 2001 . Rev Panam Salud Publica. 2007;22(1):1-11.

22. Perú, Congreso de la República. Programa Nacional Vida Digna . Decreto Supremo 006-2012-MIMP. Lima: Congreso de la República; 2012.

23. Trigás-Ferrín M, Ferreira-González L, Meijide-Míguez H. Escalas de valoración funcional en el anciano . Galicia Clin. 2012;72(1):11-6.

24. Organización Panamericana de la Salud. Evaluación funcional del adulto mayor . Washington D.C.: Organización Panamericana de la Salud.

25. Pfeiffer E. A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients . J Am Geriatr Soc. 1975 Oct;23(10):433-41.

26. Marc LG, Raue PJ, Bruce ML. Screening performance of the 15-item geriatric depression scale in a diverse elderly home care population . Am J Geriatr Psychiatry. 2008 Nov;16(11):914-21. doi: 10.1097/JGP.0b013e318186bd67.

27. Chongsuvivatwong V (2012). epicalc: Epidemiological Calculator. R package version 2.15.1.0.

28. Ruiz J, Aigneren M. Censo de adultos mayores en situación de calle, Medellín 2009 . Medellín: Universidad de Antioquia, 2009.

29. Kondratas A. Estimates and public policy the politics of numbers. Housing Policy Debate. 1991;2(3):631-47.

30. Chelimsky E. Politics, Policy Making, Data, and the Homeless . Housing Policy Debate. 1991;2(3):683-97.

31. Albala C, Lebrão ML, Leon Diaz EM, Ham-Chande R, Hennis AJ, Palloni A, et al. Encuesta Salud, Bienestar y Envejecimiento (SABE): metodología de la encuesta y perfil de la población estudiada . Rev Panam Salud Publica. 2005;17(5-6):307-22.

32. Rota-Bartelink A, Lipmann B. Causes of homelessness among older people in Melbourne, Australia . Aust N Z J Public Health. 2007 Jun;31(3):252-8.

33. Gaebel W, Zielasek J. Homeless and mentally ill - a mental healthcare challenge for Europe . Acta Psychiat Scand. 2015 Jan 24:236-8.

34. Cross W, Hayter M, Jackson D, Cleary M. Editorial: Meeting the health care needs associated with poverty, homelessness and social exclusion: the need for an interprofessional approach . J Clin Nurs. 2012 Apr;21(7-8):907-8. doi: 10.1111/j.1365-2702.2011.03945.x.

35. National Institute on Aging. Biology of Aging. Research Today for a Healthier Tomorrow . Washington D.C: National Institute of Health; 2011.

Correspondence: Marlene Moquillaza Risco

Address: Jr. Camaná 616. Lima - Perú.

E-mail: mmoquillaza@mimp.gob.pe