‘Giving a voice to the unheard’? Is female youth (15–24 years) suicide linked to restricted access to family planning? Comparing two Catholic continents

Abstract

Suicide is a neglected area of social work research and suicides are often influenced by cultural factors. WHO data on female youth (15–24 years) suicide, and Undetermined Deaths (UnD), the most likely source of under-reported suicides, are analysed from two continents sharing religio-cultural views on suicide and family planning, that is, Western European Catholic countries (WECC) and Latin American countries (LatAC). It was found that LatAC female youth suicides and UnD were significantly higher than WECC, and correlated with higher birth rates and poverty. This may indicate that restricted access to family planning contributes to more births, poverty, unwanted pregnancies and suicides.

Keywords

family planning female suicide youth

Introduction

Social work is concerned with the interaction of individuals, families and society and often identifies previously unrecognized social issues. This article seeks to give voice to another ‘voice of the unheard’ by focusing upon female youth (15–24 years) suicide victims in Latin American countries (LatAC). We explore possible under-reported suicides to alert policy-makers and practitioners, as female youth suicide is sometimes related to unwanted pregnancies (Bunevicius et al., 2009; Langer, 2002; Schmiege and Russo, 2005). In particular, we wish to explore suicide on two continents that share the same formal religo-culture views on suicide and family planning, that is, LatAC and Western European Catholic countries (WECC).

Culture influencing suicide data

A perennial problem in suicide research is the accuracy of suicide figures, taken from secondary sources, in this case from WHO annual mortality rates (WHO, 2008). One factor is the religio-cultural stigma surrounding suicide, which can influence the collection of suicide statistics (Cvinar, 2005; Kelleher et al., 1998). However, this has lessened recently so that Catholic countries such as Ireland, Italy and Spain, who used to report very low suicide rates, now report suicide more readily (Cantor et al., 1997; Levi et al., 2003).

Some have argued, however, that there is an under-reporting of suicides when coroners, seeking to save the bereaved family further distress, designate the death as an ‘open verdict’, often on the grounds that intent cannot be established. Such deaths are then formally categorized as ‘Undetermined Deaths’ (UnD). It has been asserted that such Undetermined Deaths should be included in suicide research (Linsley et al., 2001; Stanistreet et al., 2001) and the Office of National Statistics has now begun to provide combined rates of ‘Self-inflicted’ and ‘Undetermined’ deaths (ONS, 2011).

Another factor is the cultural and theological anathema against suicide and although the Roman Catholic Church urges ‘compassion and understanding’ of the suicide victim, especially when there is a ‘disordered state’, nonetheless, irrespective of motives, a suicide is still considered a ‘mortal sin’ (Roman Catholic Church, 2011). This is thought to have contributed to the disproportionately high levels of UnD found within male youth (15–24 years) deaths in many LatACs, when compared with male youth (15–24 years) deaths in Western nations (Pritchard and Hean, 2008). This theological anathema also occurs in Islamic countries, where there are often legal sanctions against suicide inhibiting people from seeking help (Cvinar, 2005) and these countries have comparatively high levels of UnD, which are thought to be under-reported suicides (Khan and Hyder, 2006; Khan et al., 2008; Okash, 2005; Pritchard and Amunallah, 2007).

Pregnancies and suicide

There is a long-standing association between depression/suicidal behaviour and unwanted pregnancies, albeit far less than previously (Bunevicius et al., 2009; Langer, 2002; Newport et al., 2007; Schmiege and Russo, 2005). However, it should be noted that wanted pregnancies can also be linked to depression (Cumming and Klein, 2007; Dave et al., 2011) and the more pregnancies a woman has, the more often bio-psychosocial problems emerge (Bunevicius et al., 2009; Cumming and Klein, 2007; Schmiege and Russo, 2005; Sedgh et al., 2007).

Family planning access

The topic is controversial but the United Nations Millennium Development Goals’ aspiration to reduce maternal peri-mortality (UNMDG Task Force, 2009), argues that the only way to achieve safer childbirth is through a rights-based approach. This is interpreted to mean that women should be the ultimate arbitrator of their own fertility, from which so many other socio-economic opportunities stem (Boama and Arulkumaran, 2009; Sedgh et al., 2007). Legislation stemming from women’s rights movements has helped Western women control their fertility via easy access to family planning (Boama and Arulkumaran, 2009), seen in the fact that currently Catholic Austria, Italy and Spain, have lower birth rates than Protestant/secular Britain (US Bureau of Statistics, 2009).

It is argued that if all women were able to control their own fertility, this would lead to fewer unwanted pregnancies and therefore fewer suicides, which has occurred in a number of Western countries (Boama and Arulkumaran, 2009; Pritchard and Hansen, 2005; WHO, 2008). Whilst in every country, ‘psychosocial’ problems are often found to be linked to unwanted pregnancies (Cheng and Horon, 2010; Sedgh et al., 2007), and in some LatACs suicide or homicide is the highest cause of death of pregnant women (Carroll, 2007a, 2007b; Langer, 2002; Sousa et al., 2010), rather than such medical complications as eclampsia.

Social attitudes in WECC on family planning have changed as all now have full access to family planning services, although in the latest countries to have access, Portugal and Ireland, these changes remain controversial (Dyer, 2010; Oliveira da Silva, 2009).

It is however, very different in the predominantly Roman Catholic Latin America, where apart from Cuba, family planning is restricted, as ‘artificial’ methods are considered a mortal sin (Vatican, 2011) and are seen as socially abhorrent (Curlin et al., 2007; Palermo et al., 2010; Reploge, 2007). Nevertheless, it has been found that many LatACs (both Central and Southern Latin America) have double the rate of Western European abortions, which of course will be mainly illegal and unsafe (Carroll, 2007a; Sedgh et al., 2007). However, Mexico City has recently introduced ‘abortion’ within first three months because of their concerns that unwanted pregnancies result in substantial numbers of unsafe abortions and deaths (Palermo et al., 2010; Sedgh et al., 2007; Tuckman, 2007).

A poverty dimension

One major difference between the two Catholic continents is likely to be poverty (Boama and Arulkumaran, 2009; Cheng and Horon, 2010; Sedgh et al., 2007) and this dimension, based upon comparative per capita Gross National Incomes (GNI) (US Bureau of Statistics, 2009) between the countries, will be briefly explored as a context in which examine the birth and mortality outcomes.

The research question

As it is female youths (15–24 years) who are the most vulnerable to unwanted pregnancies (Langer, 2002; Sedgh et al., 2007), our focus is upon female youth suicide rates that are usually the lowest amongst women (WHO, 2008). The question we seek to answer is whether there is a statistical link between female youth (15–24 years) suicide, undetermined deaths and restricted access to family planning on two continents that share the same formal Catholic doctrine on family planning and suicide (Vatican, 2011).

In order to answer the question, we contrast suicide and UnD rates for youth (15–24 years) with suicide and UnD rates for ‘all age’ (i.e. the rate for the total female population) and explore birth rates, as the higher the birth rate the more likely it is that there will be more unwanted pregnancies.

It is appreciated that to prove a direct connection between restricted access to family planning and suicide would require actual samples of cases, but this comparative epidemiological analysis can demonstrate whether such a study would be justified. The null-hypotheses are that there will be no significant statistical differences between LatACs and WECCs, with regard to:

Patterns of female all age, youth suicide and UnD rates.

An association between suicide, UnD and birth rates.

Any association between poverty (GNI) and the mortalities and birth rates.

Methodology

The latest World Health Organization (WHO, 2008) data are used for a country’s last three years (2003–5) suicide and UnD rates to compare the 17 LatACs and 10 WECCs. A few countries had earlier data than 2003–5 and these are noted in the tables. All WECCs have significant minorities of other faiths but are predominantly Roman Catholic. Greece is also included in the analysis because Orthodoxy shares the same theological view of suicide. Belgium’s data, whilst only up to 1997, are included to provide as comprehensive an analysis of WECCs as possible, and as will be seen, Belgium’s pattern of suicide and UnD are typical of other WECCs.

LatACs also have minorities of other faiths but even in Cuba, the majority of the population are Catholic (Roman Catholic Church, 2011). However, there are data from one other South American country, Guyana, but it has a very different post-colonial past, reflected in its population, with more people from a Hindu or Islamic background than Catholic tradition and is religio-culturally atypical to the LatACs, and hence not included in the study (US Library Congress, 2011).

Suicides

The strength of World Health Organization (WHO) mortality statistics is that they are collated uniformly from each member country of confirmed deaths, although they are usually five or more years behind the current calendar year. Suicides are coded X60-84 in the latest International Classification of Diseases, 10th edition and consist of ‘Intentional . . . self-poisoning . . . exposure to (range of) drugs . . . pesticides . . . vapours . . . gases . . . hanging, suffocation, drowning, self-immersion, firearms . . . smoke, steam, fire, jumping, lying before moving object, crashing motor vehicle, . . . and ‘‘unspecified means’” (WHO, 1992). There is no separate category for possible ‘assisted suicides’. The key to the suicide verdict is that ‘intent’ has been established, that is, that it was self-inflicted and intended, differentiating it from an accidental or undetermined death.

Suicide rates are reported in rates per million of population for ‘all age’ suicide rate and then per million of population in decades of age bands: 15–24 years (youths), 25–34 years, 35–44 years up to 75+. Typically, as will be seen, female youth suicide rates are the lowest among the adult age bands in every WECC (WHO, 2008).

In view of the aforementioned cultural issues surrounding suicides it is generally accepted that there will be a degree of under-reporting of suicides in most countries (Reynders et al., 2011; Violanti, 2010), hence the rational for analysing Undetermined Deaths (UnD).

‘Undetermined Deaths’ (UnD)

An undetermined death is so designated because there was ‘insufficient information to determine whether it was an accident, self-harm or an assault’ (Y10-35 and Y87-89; WHO, 1992) and no ‘intent’ could be established. However, the methods of lethality are virtually the same as those for suicide as outlined above (poisoning, firearms, drowning, etc.) and are the biggest potential source of under-reported suicide (Linsley et al., 2001; Reynders et al., 2011; Stanistreet et al., 2001; Violanti, 2010).

Poverty dimension

An important context in comparing between European and South American countries is poverty. Thus the mortality and birth rate outcomes results are placed in the context of comparisons between per capita Gross National Income (GNI), based upon purchasing power parity (ppp) in US dollars between the continents (US Bureau of Statistics, 2009). Comparative poverty is logically linked to the lowest per capita GNI, which will be correlated with birth and mortality data. Unfortunately, the most recent GNI statistics up to 2009 did not report on more than half the countries under review, hence the use of the 2006 report for the year 2000, but this still only provides information on 12 LatACs and nine WECCs. This limitation needs to be borne in mind when discussing the findings.

Chi square tests are used to compare the average mortality patterns of the WECC with each LatAC. To determine any statistical association between suicide, UnD, birth rates and poverty, that is, the lowest GNI, the Spearman Rank order (Rho) correlation is used. No birth rate or GNI data were available for the Dominican Republic and Nicaragua and no GNI data for Cuba, the Dominican Republic, Nicaragua, Panama, Paraguay and Austria who therefore are excluded from this part of the study.

Finally, we calculate an all age:youth ratio for suicide rates and UnD rates, as an indicator of the extent to which youth rates might differ from all age rates, as it would be expected that all age rates would be the higher (WHO, 2008). For example, an all age rate of 75 per million to a youth suicide rate of 50 per million would yield a ratio of 0.67, thus for every 100 all age suicides there would be 37 youth suicides.

Results

Western European Catholic countries (WECC) female suicides

To place the female youth and all age suicides in context, the average WECC suicide rates across the age bands are given here.

Average all age suicide was 63 per million (pm) population; 15–24 years (youth) 34 pm; 25–34 years 35 pm; 35–44 yrs 57 pm; 45–54 years 75 pm; 55–64 years 97 pm; 65–74 yrs 99 pm; and 75+ 147 pm, confirming that WECC female youth suicide rates are the lowest among all the age bands. Table 1 shows the three-year all age suicide and UnD rates for each WECC. Belgium was highest at 113 pm followed by Switzerland at 102 pm; the lowest was 13 pm in Greece and in Italy 33 pm.

Table 1.

Female suicide and ‘Undetermined Death’ rates per million (pm) population in Western European Catholic countries 2003–5 (ranked by widest all age:youth suicide ratio).

Country years, suicide & UnDs	All age	Youth (15–24 yrs)	All age:youth ratios
1. Ireland 2003–5	39	43	1.10
Undetermined	8	4	0.50
UnD:suicide ratio	1:4.9	1:10.8	1:2.2
2. Portugal 2001–3	44	47	1.07
Undetermined	26	7	0.22
UnD:suicide ratio	1:1.7	1:6.7	1:3.9
3. Greece 2003–5	13	7	0.54
Undetermined	0	0	0.00
UnD:suicide ratio	xxx	xxx	xxx
4. Italy 2001–3	33	17	0.52
Undetermined	8	3	0.38
UnD:suicide ratio	1:4.1	1:5.7	1:1.4
5. Belgium 1995–7	113	58	0.51
Undetermined	15	11	0.73
UnD:suicide ratio	1:7.5	1:5.3	1:0.7
6. Spain 2003–5	39	19	0.49
Undetermined	1	1	1.00
UnD:suicide ratio	1:39	1:19	1:0.5
7. Austria 2003–5	78	37	0.47
Undetermined	14	9	0.64
UnD:suicide	1:5.6	1:4.1	1:0.7
8. Switzerland 2003–5	110	51	0.46
Undetermined	44	12	0.27
UnD:Suicide ratio	1:2.5	1:4.3	1:1.7
= 9. Germany 2003–5	63	24	0.38
Undetermined	23	6	0.26
UnD:suicide ratio	1:2.7	1:4.0	1:1.5
= 9. France 2003–5	92	35	0.38
Undetermined	3	3	1.00
UnD:suicide ratio	1:30.7	1:11.7	1:0.4
Average suicide	63	34	0.54
Average undetermined	14	6	0.43
UnD:suicide ratio	1:4.5	1:5.7	1:1.3
Combined	77	40	0.52

All age suicide rates were mainly higher than youth suicide rates, but Ireland and Portugal, the latest countries to have access to family planning, had youth suicide rates just slightly higher than all age suicide rates, but this was significantly different from the other WECCs (p < 0.001).

Average WECC suicides were all age 63 pm and youth 34 pm. The youth:all age ratio was 0.54, so for every 100 all age suicides there would be 46 youth suicides.

WECC ‘Undetermined Deaths’ (UnD)

Average WECC UnD rates across the age bands were: all age 14 pm, youth 6 pm; 25–34 years 7 pm; 35–44 years 9 pm; 45–54 years 11 pm; 55–64 years 13 pm; 65–74 years 24 pm; and 75+ 108 pm, similar to the age pattern for WECC suicide rates. The highest all age UnD rates were 44 pm in Switzerland and in Portugal 26 pm. The lows were 1 pm in Greece and Spain. All age UnD rates were higher than youth UnD rates in every country. All age UnD rates averaged 14 pm compared to 6 pm for female all age:youth ratio of 0.43. It is noteworthy in contrasting UnD to suicides that average all age and youth UnD:suicide ratios are 1:4.5 and 1:5.7 respectively.

Latin American countries (LatAC) suicides

Table 2 lists LatAC suicides. The average LatAC suicide rates across the age bands were: all age at 28 pm; youth 48 pm; 25–34 years 35 pm; 35–44 years 37 pm; 45–54 years 44 pm; 55–64 years 63 pm; 65–74 years 52 pm; and 75+ 47 pm. Unlike the average WECC, LatAC youth average suicide rates were higher than most other age bands.

Table 2.

Latin American countries suicide and UnDs 2003–5 rates per million (pm) population (ranked by widest all age:youth ratios).

Country, years, suicides & UnDs	All age	Youth(15–24 yrs)	All age:youth ratios
1. Peru 1998–200l	6	26	4.33
Undetermined	27	69	2.56
2. El Salvador 2001–3	39	110	2.82
Undetermined	0	0	0.00
3. Ecuador 2003–5	37	104	2.81
Undetermined	9	8	2.00
4. Paraguay 2002–4	22	60	2.73
Undetermined	11	29	2.64
5. Nicaragua 2003–5	35	95	2.71
Undetermined	1	2	2.00
6. Colombia 2002–4	23	62	2.69
Undetermined	21	23	1.10
7. Costa Rica 2003–5	22	44	2.30
Undetermined	7	7
8. Guatemala 2002–4	10	21	2.10
Undetermined	107	136	1.27
9. Panama 2002–4	19	36	1.89
Undetermined	3	5	1.67
10. Mexico 2003–5	13	23	1.77
Undetermined	8	9
11. Venezuela 2003–5	15	26	1.73
Undetermined	47	37	0.79
12. Argentina 2003–5	34	50	1.47
Undetermined	18	19	1.06
13. Chile 2003–5	32	43	1.34
Undetermined	1	1	1.00
14. Brazil 2003–5	18	23	1.28
Undetermined	26	14	0.52
15. Uruguay 2003–5	58	57	0.98
Undetermined	4	2	0.50
16. Cuba 2003–5	65	40	0.62
Undetermined	12	5	0.41
17. Dominican Rep 2002–4	24	6	0.79
Undetermined	62	49	0.76
Average suicide	28	48	1.71
Average UnD	21	25	1.19
UnD:suicide ratio	1:1.3	1: 1.9	1.46
Average combined	49	73	1.49

Cuba 65 pm and Uruguay 58 pm had the highest all age suicide rates compared to lows of 6 pm in Peru, 10 pm in Guatemala and 13 pm in Mexico. Youth suicide rates spanned from 110 pm in El Salvador and 104 pm in Ecuador to lows of 6 pm in the Dominican Republic, 21 pm Guatemala and 23 pm in Brazil. The average LatAC all age suicide rate was 28 pm but the average youth (15–24 years) suicide rate was 60 pm, with an all age:youth ratio of 1.71. Thus for every 100 all age suicides there were 171 youth suicides.

Apart from Cuba, the Dominican Republic and Uruguay, the other LatAC youth suicide rates were often substantially higher than all age suicide rates.

LatAC ‘Undetermined Deaths’ (UnD)

The average LatAC UnD rates (Table 2) were less stable as there were marked differences between the countries. Nine countries had rates less than 12 pm and six had rates ranging from 24 pm to 107 pm. Average rates across the age bands were all age 21 pm and youth 25 pm; 25–34 years 22 pm; 35–44 years 24 pm; 45–54 years 25 pm; 55–64 years 48 pm; 65–74 years 67 pm; and 75+ 70 pm. Youth rates were higher than half the other age bands and it may be the higher older >55 age-band UnDs were also a source of ‘hidden’ suicides for women likely to have been mothers and grandmothers.

All age UnD rates ranged from 107 pm in Guatemala and 62 pm in the Dominican Republic, down to lows of 1 pm in Chile and Nicaragua and none in El Salvador. The youth UnD rate ranged from 136 pm in Guatemala and 69 pm in Peru to lows of 1 pm in Chile and 2 pm in Nicaragua and Uruguay. Unlike the WECCs, many LatAC youth UnD rates exceeded the all age UnD rates. The average LatAC all age UnD rate was 21 pm compared with average youth UnD rate at 25 pm. The youth:all age ratio was 1.19. Thus for every 100 all age UnD there were 119 youth undetermined deaths. The youth and all age UnD:suicide ratios are 1:1.3 and 1:1.9 respectively, again markedly different from WECC patterns.

Comparing LatAC versus WECC suicides and UnDs

Table 3 shows the chi square test results of LatAC, patterns of suicide and UnD rates compared with WECC rates. Except for Cuba and the Dominican Republic every other LatAC pattern of suicide to UnD rates were significantly different from the WECC, with especially marked differences in Colombia, Costa Rica, El Salvador, Ecuador, Nicaragua and Peru (p < 0.0001) and involving both Central and Southern LatACs.

Table 3.

Comparing LatAC with average WECC patterns of suicide and UnD rates (chi square and p values).

Country	Suicide χ² p value	Combined suicide & UnD
Argentina	10.84 - <0.01	12.49 <0.001
Brazil	5.2651 - <0.05	2.6592 n.sig
Chile	8.4927 - <0.01	9.9668 <0.01
Colombia	26.09 - <0.0001	24.67 <0.0001
Costa Rica	18.73 - <0.0001	16.71 <0.001
Cuba	0.0201 - n.sig	0.6383 n.sig
Dominican Rep.	2.4059 - n.sig	0.6338 n.sig
El Salvador	36.39 - <0.0001	41.87 <0.0001
Ecuador	35.34 - <0.0001	41.52 <0.0001
Guatemala	10.24 - <0.01	31.22 <0.0001
Mexico	8.9155 - <0.01	10.2471 <0.001
Nicaragua	32.74 - <0.0001	37.72 <0.0001
Panama	13.08 - <0.001	15.78 <0.001
Paraguay	25.89 - <0.0001	36.12 <0.0001
Peru	20.64 - <0.0001	39.59 <0.0001
Uruguay	4.5241 - <0.05	5.1994 <0.03
Venezuela	7.1557 - <0.01	6.4973 <0.03

Birth rates and poverty (GNI)

Table 4 shows birth rates and per capita GNI for the countries under review. The WECC birth rates ranged from 14.2 per thousand (pt) women (16–44 years) in Ireland to 8.3 pt in Germany, averaging 9.6 pt, with Ireland being one Standard Deviation (SD) above the mean (Table 4). In the 15 LatACs, birth rates ranged from 30.4 pt in Paraguay down to 11.9 pt in Cuba and 15.4 pt in Chile, averaging 20.9 pt, more than double the WECC average.

Table 4.

LatAC & WECC current birth rates per 1000 women (15–44 years) and per capita $GNI (purchasing power parity ranked by highest birth rate).

LatAC& rank	Birthrate	Per capita$GNI	WECC &rank	Birth rate	Per capita $GNI
1. Paraguay	30.4	n/a	Ireland	14.2	22,520
2. Guatemala	30.2	3770	Belgium	12.4	27,470
3. El Salvador	26.5	4410	France	12.2	24,420
4. Panama	23.5	n/a	Portugal	11.00	16,990
5. Ecuador	22.7	2910	Switzerland	10.9	35,450
6. Mexico	21.0	8790	Spain	10.1	19,260
7. Peru	20.9	4660	Austria	9.7	N/a
8. Colombia	20.8	6060	Greece	9.2	16,860
9. Costa Rica	19.2	7980	Italy	9.1	23,470
10. Venezuela	18.9	5740	Germany	8.3	24,920
11. Argentina	16.9	12,050
12. Brazil	16.8	7300
13. Uruguay	16.7	8880
14. Chile	15.4	9100
15. Cuba	11.9	n/a
Nicaragua	n/a	n/a
Dominican Rep.	n/a
Range average birth rate	11.9–30.420.9			8.3–14.210.7

The nine WECCs’ per capita GNI ranged from $16,860 in Greece to $35,450 in Switzerland, averaging $23,484. This was in stark contrast to the 12 LatACs whose GNI ranged from $2910 in Ecuador to $12,050 in Argentina, averaging $6804, less than a third of the WECCs’.

Correlating birth rates, mortalities and lowest GNI

Listed in Table 5 are the correlations of births, poverty and mortalities. For the WECCs, in regard to birth rates and youth:all age suicide ratios they were significantly positively correlated (p < 0.025), as were lowest GNI and youth:all age ratios (p < 0.025), and though all age mortalities and birth rates were not significantly linked, youth deaths and births were (p < 0.05), suggesting an association between combined youth deaths and birth rate. However, in respect to poverty or lowest GNI per capita and WECC mortalities, whilst WECC GNI and birth rates were not significantly correlated, there were significant but negative correlations between GNI and youth deaths (p < 0.05) and all age deaths (p < 0.005).

Table 5.

Spearman rank order correlations of mortalities, birth rates and lowest $GNI (purchasing power parity).

Correlation	Rho	p value
Western European CC
Birth rates & all age:youth ratios	+ 0.7061	<0.025
GNI lowest & all age:youth ratios	+ 0.7545	<0.025
Birth rates & combined youth deaths	+ 0.5848	<0.05
Birth rates & combined all age deaths	+ 0.2080	n.s.
GNI lowest & combined youth deaths	− 0.5958	<0.05
GNI lowest & combined all age deaths	− 0.8667	<0.005
GNI lowest & birth rates	− 0.1333	n.s.
Latin America
Birth rates & all age:youth ratios	+ 0.7839	<0.001
GNI lowest & all age:youth ratios	+ 0.6993	<0.025
Birth rates & combined youth death rates	+ 0.5321	<0.025
Birth rates & combined all age death rates	− 0.1125	n.s.
Birth rates & lowest GNI	+ 0.7832	<0.005
GNI lowest & combined youth deaths	+ 0.6188	<0.025
GNI lowest & combined all age deaths	+ 0.2604	n.s.

For the LatACs, the strongest correlations were between births rates and all age:youth ratios, which were highly significant (p < 0.001), followed by the lowest GNI and all age:youth ratios (p < 0.025), and lowest GNI and birth rates (p < 0.005).

Both birth rates and lowest GNI were positively significantly correlated with youth deaths (p < 0.025). Conversely, births and lowest all age deaths were not significantly linked. This strongly suggests an interaction of higher birth rates and poverty with youth suicides and UnDs.

Discussion

The limitations of the study were that Ireland and Portugal’s mortality rates, the latest WECCs to have easy access to family planning, were significantly different from the other countries and this slightly reduces the average differences found between the two continents. Also, there was no exact temporal matching across continents and birth data were unavailable for the Dominican Republic and Nicaragua. There was also an absence of GNI data for seven LatACs and one WECC. The key limitation, however, is that only a clinically based study could demonstrate a direct association of whether actual suicides were influenced by restricted access to family planning and poverty, rather than this aggregated analysis based upon secondary data. Nonetheless, we have shown unequivocally that LatAC female youth have higher rates of suicide and undetermined deaths than in WECCs, as well as almost double the birth rates and worse poverty, strongly suggesting a link with limited access to family planning. This result should be an impetus for an appropriate study of a representative sample from the countries under review.

Main findings

The null hypothesis that there would be no significant statistical difference in the patterns of all age and youth suicide and UnD rates between the two continents can be rejected. Also rejected is the hypothesis that there would be no difference between mortalities and birth rates, as there are markedly different patterns in the two continents.

It is noteworthy that Cuba, which had the lowest LatAC birth rate and the fifth lowest combined youth suicide and UnD rates, was only one of two LatACs (the other being the Dominican Republic) that was not significantly different from the mortality patterns in the WECC.

The third hypothesis is also rejected, as there is a strong association in regard to poverty, having the lowest per capita GNI and deaths in the LatACs, especially strong for youths, although the strongest correlation was between birth rates and all age:youth mortality ratios, which indicates the interaction of more pregnancies and higher suicides amongst LatAC female youth.

In most of the Latin countries under review, family planning is relatively restricted, making it more likely that female youths (15–24 years) are more vulnerable to unwanted pregnancies (Langer, 2002; Sedgh et al., 2007), compounded by poverty, and, in a response to their situation, they are at greater risk of being involved in suicidal behaviour. Should this occur, it appears that because of the cultural pressures and stigmas surrounding suicide, there is a greater likelihood of the death being categorized as ‘undetermined’, in part to save the family further distress (Linsley et al., 2001; Stanistreet et al., 2001; Violanti, 2010), leading to higher levels of UnD, which is the basis for the argument that UnDs are a likely source of under-reported youth suicides. This also appears to be the case in Islamic countries (Khan and Hyder, 2006; Khan et al., 2008; Pritchard and Amunallah, 2007; Vakili et al., 2010) as well as in the majority of LatACs. Of course, this can never be known, as an undetermined death might be just that, namely, there was insufficient evidence to determine the cause of death. There is, however, some indirect support for our interpretation from the important study of Sedgh et al. (2007), which found that abortions in Central and South America were double those in Western Europe and these would be both illegal and unsafe (see also Kulczycki, 2007; Sousa et al., 2010).

The pursuit of women’s rights in Western countries has led to relatively easy access to full family planning, although this continues to be challenged (Dyer, 2010; Oliveira de Silva, 2009). LatACs have double the WECC birth rates as women are less able to effectively control their own fertility (Carroll, 2007a; Sousa et al., 2010). Furthermore, higher birth rates continue to be associated with child poverty, maternal mortality, violence, exploitation and oppression of women in many countries, but especially those in which women’s rights are less established (e.g. Boama and Arulkumaran, 2009; Lehrer et al., 2009; Sedgh et al., 2007; Sousa et al., 2010; Vakili et al., 2010).

There is a poverty dimension. The links between the mortalities and the lowest GNIs in the LatACs were not a surprise and appear to be an additional problem for female LatAC youth. It is noteworthy, however, that in the WECCs, there was a significant inverse relationship between poverty and the mortalities, certainly highlighting the different patterns found in Latin America, but this will require further country-specific research to explain these counter-intuitive results.

It is argued that until the ‘voices of the unheard’ in many Latin American countries are listened to, there is unlikely to be the necessary shift in cultural attitudes towards women’s rights, and young women, faced with an unplanned pregnancy, will continue to find themselves in an unsupportive and uncompassionate environment. This poses a special ethical challenge for social work, a discipline that advocates anti-oppressive practice but clashes with our responsibility to be culturally sensitive. From a traditional Catholic perspective, family planning is morally wrong (Vatican, 2011), yet to achieve the UN Millennium Development Goal of safer childbirth the issue has to be placed in a ‘rights context’ (UNMDG Task Force, 2009). What should or can Latin American social work colleagues do? On the one hand, it should be remembered how relatively recent is the women’s rights movement in the West (e.g. Greer, 1976). In a ‘globalized’ world it is hoped that by providing objective evidence these human dilemmas might be recognized, so that the ‘voices of the unheard’ might be listened to and Latin America can begin to reduce the tragic toll of female youth suicide.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Author biographies

Colin Pritchard is Research Professor in Psychiatric Social Work, School of Health & Social Care, Bournemouth University and Visiting Professor, Department of Psychiatry, University of Southampton, UK.

Sion Roberts is a Consultant Psychiatrist at The Amy Evans Centre, Barry, Wales, UK.

Claire E. Pritchard is a Senior Social Worker in the London Borough of Islington, UK.

References

Boama

Arulkumaran

(2009) ‘Safer Childbirth: A Rights-based Approach’, International Journal of Gynaecology & Obstetrics 106: 125–7.

Bunevicius

Kusminskas

Bunevicius

Pop

V.J.

(2009) ‘Psychosocial Risk Factors for Depression during Pregnancy’, Acta Obstetricia et Gynecologica Scandinavica 88: 599–605.

Cantor

Leenaars

Lester

(1997) ‘Under-reporting of Suicide in Ireland 1960–89’, Archives of Suicide Research 3: 5–12.

Carroll

(2007a) ‘Pope Warns Latin America against Marx’, Guardian, 15 May, p. 14.

Carroll

(2007b) ‘Killer Law’, Guardian, 10 October, pp. 4–7.

Cheng

Horon

(2010) ‘Intimate-Partner Homicide among Pregnant and Postpartum Women’, Obstetrics & Gynaecology 115: 1181–6.

Cumming

G.P.

Klein

(2007) ‘The Emotional Burden of Miscarriage for Women and their Partners: Trajectories of Anxiety and Depression over 13 Months’, British Journal of Obstetrics & Gynaecology 114: 1138–45.

Curlin

Lawrence

Chin

Lantos

(2007) ‘Religion, Conscience and Controversial Clinical Practices’, New England Journal of Medicine 356: 593–600.

Cvinar

(2005) ‘Do Suicide Survivors Suffer Social Stigma? A Review of the Literature’, Perspectives in Psychiatric Care 41: 14–21.

10.

Dave

Petersen

Sherr

Nazareth

(2011) ‘Incidence of Maternal and Paternal Depression in Primary Care: A Cohort Study Using a Primary Care Data Base’, Archives of Paediatric & Adolescent Medicine 6, Epub ahead of print.

11.

Dyer

(2010) ‘Women Challenge Irish Law on Abortion in Europe on Human Rights Grounds’, British Medical Journal 339: 5457.

12.

Greer

(1976) The Female Eunuch. London: Flamingo Press.

13.

Kelleher

M.J.

Chambers

Corcoan

Williamson

E.T.

Keeley

H.S.

(1998) ‘Religious Sanctions and Worldwide Suicide’, Crisis 19: 78–86.

14.

Khan

Nacqi

Thaver

Prince

(2008) ‘Epidemiology of Suicide in Pakistan: Determining Rates in Six Cities’, Archives of Suicide Research 12: 155–60.

15.

Khan

M.M.

Hyder

A.A.

(2006) ‘Suicides in the Developing World: A Case Study from Pakistan’, Suicide and Life Threatening Behaviour 36: 76–81.

16.

Kulczycki

(2007) ‘The Abortion Debate in Mexico: Realities and Stalled Policy Reform’, Bulletin of Latin American Research 26: 50–68.

17.

Langer

(2002) ‘Unwanted Pregnancy: Impact on Health and Society in Latin America and the Caribbean’, Revista Panamericana de Saluda Pública 11: 192–204.

18.

Lehrer

E.L.

Lehrer

Krauss

R.C.

(2009) ‘Religion and Intimate Partner Violence in Chile: Macro–Micro Level Influence’, Social Science Research 38: 635–43.

19.

Levi

La Vecchia

Lucchini

(2003) ‘Trends in Mortality from Suicide 1969–99’, Acta Psychiatrica Scandinavica 108: 341–9.

20.

Linsley

K.R.

Schaparia

Kelly

T.P.

(2001) ‘Open Verdicts versus Suicide: Importance to Research’, British Journal of Psychiatry 178: 465–8.

21.

Newport

Levy

Pennell

(2007) ‘Suicidal Ideation in Pregnancy: Assessment and Clinical Implications’, Archives of Women’s Mental Health 10: 181–7.

22.

ONS (2011) National Statistics. London: Office of National Statistics (www.ons.org.uk).

23.

Okash

(2005) ‘Mental Health in Egypt’, International Journal of Psychiatry and Relative Science 20: 917–33.

24.

Oliveira da Silva

(2009) ‘Reflections on the Legislation of Abortion in Portugal’, European Journal of Reproductive Health Care 14: 245–8.

25.

Palermo

T.M.

Wilson

K.S.

Garcia

Diaz-Olavarrieta

(2010) ‘Abortion and Women’s Roles in Society, Opinions from Tlaxcala Mexico’, Salud Públicade México 52: 52–60.

26.

Pritchard

Amunallah

(2007) ‘An Analysis of Suicide and ‘‘Undetermined’’ Deaths in 17 Predominantly Islamic Countries Contrasted with the UK’, Psychological Medicine 36: 1–10.

27.

Pritchard

Hansen

(2005) ‘Adolescent and Youth Suicide in England and Wales: An International Comparison’, International Journal of Adolescent Medicine & Health 17: 239–53.

28.

Pritchard

Hean

(2008) ‘Suicide and ‘‘Undetermined’’ Deaths among Youths and Young Adults in Latin America: Comparison with the 10 Major Developed Countries: A Source of ‘‘Hidden’’ Suicides?’, Crisis 29: 145–53.

29.

Reploge

(2007) ‘Abortion Debate Heats Up in Latin America’, Lancet 370: 305–06.

30.

Reynders

Schneider

Van Audenhove

(2011) ‘The Reliability of Suicides Rates: An Analysis of Railway Suicides from Two Sources in Fifteen European Countries’, Journal of Affective Disorders, Epub ahead of print.

31.

Roman Catholic Church (2011) ‘The Roman Catholic Catechism’. Available online at: www.roman-catholic-catechism.com.

32.

Schmiege

Russo

(2005) ‘Depression and Unwanted First Pregnancy: Longitudinal Cohort’, British Medical Journal 331: 1303.

33.

Sedgh

Handier

Singh

Ahman

Shah

(2007) ‘Induced Abortion: Estimated Rates and Trends Worldwide’, Lancet 370: 1338–45.

34.

Sousa

Lozano

Gakidou

(2010) ‘Exploring the Determinants of Unsafe Abortion: Improving the Evidence Base in Mexico’, Health Policy and Planning 25: 300–10.

35.

Stanistreet

Taylor

Jeffrey

Gabby

(2001) ‘Accident or Suicide? Predictors of Coroners’ Decisions in Suicide and Accident Verdicts’, Medicine Science & Law 41: 111–15.

36.

Tuckman

(2007) ‘Campaigners Celebrate as Mexico City Legalises Abortion on Demand’, Guardian, 26 April, p. 18.

37.

United Nations Millennium Development Goals (UNMDG) Task Force (2009) Millennium Development Goals: A Report. New York: United Nations.

38.

US Bureau of Statistics (2009) Annual Statistics of the USA. Washington, DC: USBS.

39.

US Library Congress (2011) ‘Religion: Guyana’. Available online at: http://countrystudies.us/guyana/37.htm/.

40.

Vakili

Nadrian

Fathipoor

Morowatishabad

(2010) ‘Prevalence and Determinants of Intimate Partner Violence against Women in Kazeroon, Islamic Republic of Iran’, Violence 25: 116–27.

41.

Vatican (2011) ‘The Catechism of the Catholic Church’. Available online at: www.vatican.va.

42.

Violanti

J.M.

(2010) ‘Suicide or Undetermined? A National Assessment ofPolice Suicide Death Classification’, International Journal of Emergency Mental Health 12: 89–94.

43.

WHO (1992) International Classification of Diseases, 10th edn. Geneva: World Health Organization.

44.

WHO (2008) Annual World Health Statistics. Geneva: World Health Organization (www.whowhosis.org).